Emerging Infectious Diseases

Emerging
Vol. 1, No. 1, January–March 1995
Infectious Diseases
Emerging Infections:
Getting Ahead of the Curve David Satcher
Factors in the Emergence of
Infectious Diseases Stephen S. Morse
Cat-Scratch Disease Russell Regnery and

Jordan Tappero
Barmah Forest Virus Michael D. A. Lindsay
Shigella sonnei Infection J. A. Frost
Lyme Disease Richard C. Russell
Morbillivirus Pneumonia of Horses Keith Murray
Electronic Communication and T. Demetri Vacalis

Public Health Surveillance
Communicable Diseases Helen Longbottom

Intelligence
DxMONITOR William Hueston
WHO Scientific Working Group Fred C. Tenover and
James M. Hughes
U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES

Public Health Service
Editors Liaison Representatives
Editor Anthony I. Adams, M.D. Joseph Losos, M.D.
Joseph E. McDade, Ph.D. Chief Medical Adviser Director General
National Center for Infectious Diseases Commonwealth Department of Laboratory Center for Disease Control
Centers for Disease Control Human Services and Health Ontario, Canada
and Prevention (CDC) Canberra, Australia
Atlanta, Georgia, USA Gerald L. Mandell, M.D.
David Brandling-Bennett, M.D. Liaison to Infectious Diseases Society
Perspectives Editor Director, Division of Communicable of America
Stephen S. Morse, Ph.D. Diseases University of Virginia Medical Center
The Rockefeller University
Pan American Health Organization Charlottesville, Virginia, USA
New York, New York, USA
World Health Organization
Washington, D.C., USA Robert Shope, M.D.
Synopses Editor
Phillip J. Baker, Ph.D. Director, Yale Arbovirus Research Unit
Division of Microbiology and Richard A. Goodman, M.D., M.P.H. Yale University School of Medicine
Infectious Diseases Editor, MMWR New Haven, Connecticut, USA
National Institute of Allergy and Centers for Disease Control
Infectious Diseases and Prevention (CDC) Bonnie Smoak, M.D.
National Institutes of Health (NIH) Atlanta, Georgia, USA Chief, Dept of Epidemiology
Bethesda, Maryland, USA Division of Preventive Medicine
William Hueston, D.V.M., Ph.D Walter Reed Army Institute of Research
Dispatches Editor Acting Leader, Center for Animal Washington, D.C., USA
Stephen Ostroff, M.D. Health Monitoring
National Center for Infectious Diseases Centers for Epidemiology and Robert Swanepoel, B.V.Sc., Ph.D.
Centers for Disease Control Animal Health Head, Special Pathogens Unit
and Prevention (CDC) Veterinary Services, Animal and Plant National Institute for Virology
Atlanta, Georgia, USA Health Inspection Service Sandrinham 2131, South Africa
U.S. Department of Agriculture
Managing Editor Fort Collins, Colorado, USA
Polyxeni Potter, M.A.
Roberto Tapia, M.D.
National Center for Infectious Diseases Director General de Epidemiología
James LeDuc, Ph.D. Dirección General de Epidemiología
Centers for Disease Control
and Prevention (CDC)
Advisor for Arboviral Diseases Secretaría de Salud
Atlanta, Georgia, USA Division of Communicable Diseases México
World Health Organization
Geneva, Switzerland Morag Timbury, M.D.
Director
Central Public Health Laboratory
London, United Kingdom
Emerging Infectious Diseases

Emerging Infectious Diseases is published four times a year by the National Center for Infectious Diseases, Centers for
Disease Control and Prevention (CDC), 1600 Clifton Road., Mailstop C-12, Atlanta, GA 30333, USA. Telephone
404-639-3967, fax 404-639-3039, e-mail EIDEDITOR@CIDOD1.EM.CDC.GOV.
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of CDC or the
institutions with which the authors are affiliated.
All material published in Emerging Infectious Diseases is in the public domain and may be used and reprinted
without special permission; proper citation, however, is appreciated.
Use of trade names is for identification only and does not imply endorsement by the Public Health Service or by the
U.S. Department of Health and Human Services.
Editorial Policy and Call for Articles
The goals of Emerging Infectious Diseases (EID) are to promote the recognition of new and reemerging infectious
diseases and to improve the understanding of factors involved in disease emergence, prevention, and elimination. EID
has an international scope and is intended for professionals in infectious diseases and related sciences. We welcome
contributions from infectious disease specialists in academia, industry, clinical practice, and public health as well as
from specialists in economics, demography, sociology, and other disciplines whose study elucidates the factors
influencing the emergence of infectious diseases.
EID will be published in English and will feature three types of articles: Perspectives, Synopses, and Dispatches. The
purpose and requirements of each type of article are described in detail below.
Instructions to Authors
Editorial Material: Manuscripts should be prepared to reviews of new and emerging diseases; however, timely
according to the “Uniform Requirements for Manuscripts updates of other diseases or topics are also welcome.
Submitted to Biomedical Journals” (JAMA 1993:269[17]: Synopses should be approximately 3,500 words and
2282-6). should include references, not to exceed 40. The section
Begin each of the following sections on a new page and should begin with an introduction outlining the
in this order: title page, abstract, text, acknowledgments, relationship of the issues discussed in the paper to the
references, each table, figure legends, and figures. On the emergence of infectious diseases. Use of additional
title page, give complete information about each author subheadings in the main body of the text is
(full names and highest degree). Give current mailing recommended. If detailed methods are included, a
address for correspondence (include fax number and separate section on experimental procedures should
e-mail address). Follow Uniform Requirements style for immediately follow the body of the text. Selective use of
references. Consult List of Journals Indexed in Index illustrations is encouraged. Provide a short abstract of no
Medicus for accepted journal abbreviations. Tables and more than 150 words and a brief biographical sketch.
figures should be numbered separately (each beginning Dispatches: Provide brief updates on trends in
with 1) in the order of mention in the text. Double-space infectious diseases or infectious disease research.
everything, including the title page, abstract, references, Dispatches (1,000 to 1,500 words of text) should be in a
tables, and figure legends. Italicize scientific names of “letter to the editor” format and should not be divided
organisms from species name all the way up, except for into sections. Dispatches should begin with a brief
vernacular names (viruses that have not really been introductory statement about the relationship of the topic
speciated, such as coxsackievirus and hepatitis B; to the emergence of infectious diseases. Include methods
bacterial organisms, such as pseudomonads, development; references, not to exceed five; and figures
salmonellae, and brucellae). or illustrations, not to exceed two.
Perspectives: Contributions to the Perspectives All articles will be reviewed by independent reviewers.
section are welcome from scientists and professionals in The Editor reserves the right to edit articles for clarity
all disciplines and should address factors known to and to modify the format to fit the publication style of
contribute to the emergence of infectious diseases, Emerging Infectious Diseases.
including microbial adaption and change; human Send documents in hardcopy (Courier 10-point font),
demographics and behavior; technology and industry; on diskette, or by e-mail. Acceptable electronic formats
economic development and land use; international travel for text are ASCII, WordPerfect, AmiPro, DisplayWrite, MS
and commerce; and the breakdown of public health Word, MultiMate, Office Writer, WordStar, or Xywrite.
measures. Articles should be approximately 3,500 words Send graphics documents in Corel Draw, Harvard
and should include references, not to exceed 40. The Graphics, Freelance, .TIF (TIFF), .GIF (CompuServe),
section should begin with an introduction outlining the .WMF (Windows Metafile), .EPS (Encapsulated
relationship of the issues discussed in the paper to the Postscript), or .CGM (Computer Graphics Metafile). The
emergence of infectious diseases. Use of additional preferred font for graphics files is Helvetica. If possible,
subheadings in the main body of the text is convert Macintosh files into one of the suggested formats.
recommended. If detailed methods are included, a Submit photographs in camera-ready hardcopy.
separate section on experimental procedures should
Send all manuscripts and correspondence to the
immediately follow the body of the text. Photographs and
Editor, Emerging Infectious Diseases, National Center for
illustrations are optional. Provide a short abstract (150
Infectious Diseases, Centers for Disease Control and
words) and a brief biographical sketch.
Prevention, 1600 Clifton Road, Mailstop C-12, Atlanta,
Synopses: Submit concise reviews of infectious GA 30333, USA or by e-mail on the Internet to
diseases or closely related topics. Preference will be given EIDEDITOR@CIDOD1.EM.CDC.GOV.
Contents
Emerging Infectious Diseases

Volume 1 • Number 1 January–March 1995
Perspectives
Emerging Infections: Getting Ahead of the Curve 1 David Satcher
Factors in the Emergence of Infectious Diseases 7 Stephen S. Morse
Synopsis
Unraveling Mysteries Associated with Cat-Scratch Disease, 16 Russell Regnery and
Bacillary Angiomatosis, and Related Syndromes Jordan Tappero
Dispatches
Emergence of Barmah Forest Virus in Western Australia 22 Michael D.A. Lindsay, Cheryl A.
Johansen, Annette K. Broom,
David W. Smith, and John S.
Mackenzie
An Outbreak of Shigella sonnei Infection Associated with 26 J.A. Frost, M.B. McEvoy, C.A.
Consumption of Iceberg Lettuce Bentley, Y. Andersson, and B. Rowe
?Lyme Disease in Australia—Still To Be Proven! 29 Richard C. Russell
A Novel Morbillivirus Pneumonia of Horses and its 31 Keith Murray, Russell Rogers, Linda
Transmission to Humans Selvey, Paul Selleck, Alex Hyatt,
Allan Gould, Laurie Gleeson, Peter
Hooper, and Harvey Westbury
Commentary
Electronic Communication and the Future of 34 T. Demetri Vacalis,
International Public Health Surveillance Christopher L. R. Bartlett, and
Cheryl G. Shapiro
News and Notes
Communicable Diseases Intelligence 36 Helen Longbottom
DxMONITOR: Compiling Veterinary Diagnostic Laboratory Results 36 William Hueston
WHO Scientific Working Group on Monitoring and Management of 37 Fred C. Tenover and
Bacterial Resistance to Antimicrobial Agents James M. Hughes
Electronic Access to Emerging Infectious Diseases

If you have Internet access, you can retrieve Access Methods EID-TOC sends the table of contents and
the journal electronically through file trans- allows you to request individual articles.
fer protocol (FTP), electronic mail, or World
FTP
Download the journal through anonymous To subscribe to a list, send an e-mail to
Wide Web (WWW). lists@list.cdc.gov with the following in the
FTP at ftp.cdc.gov. The files can be found in
The journal is available in three file formats: the pub/EID directory in each of the file types body of your message: subscribe listname
ASCII, Adobe Acrobat (.pdf), and PostScript listed above. (e.g., subscribe EID-ASCII).
(.ps). The ASCII version of the journal does Once you have requested a subscription, you
not contain figures. Both the .pdf and .ps files, LISTSERVer (E-Mail Lists)
will receive further instructions by e-mail.
however, contain graphics and figures and Have the journal sent to your electronic mail-
are true representations of the hard copy of box by subscribing to one of our mailing lists. WWW—Mosaic Access
the journal. The Adobe Acrobat format re- Choose the list with the file type that best Launch your WWW Mosaic software for the
quires an Adobe Reader (available at no meets your computing environment. Four Internet and connect to the following address:
charge from CDC through FTP, LISTSERVer, lists are available: http://www.cdc.gov.
or WWW). This reader is available in DOS, EID-ASCII sends the journal in ASCII format. Your WWW software will allow you to view,
Windows, UNIX, and Macintosh versions and EID-PDF sends the journal in Adobe Acrobat print, and retrieve journal articles.
will allow you to view and print the journal format. You can get the Adobe Acrobat Reader For more information about receiving Emerg-
just as it looks in hard copy. Installation in- free by subscribing to this list. ing Infectious Diseases electronically, send an
structions come with the Adobe software. e-mail to eidhelp@cidod1.em.cdc.gov.
EID-PS sends the journal in PostScript format.
Perspectives
Emerging Infections:
Getting Ahead of the Curve
David Satcher, M.D., Ph.D.
Centers for Disease Control and Prevention, Atlanta, Georgia, USA
The early history of infectious diseases was characterized by sudden, unpredictable

outbreaks, frequently of epidemic proportion. Scientific advances in the late 19th and
early 20th centuries resulted in the prevention and control of many infectious diseases,
particularly in industrialized nations. Despite these improvements in health, out-
breaks of infectious disease continue to occur, and new infections emerge. Since 1987,
the National Academy of Science’s Institute of Medicine (IOM) has published three
reports that have identified erosion of the public health infrastructure among the
factors contributing to new and reemerging infectious diseases. In partnership with
many public and private organizations in the United States and abroad, the Centers
for Disease Control and Prevention (CDC) has developed a strategic plan that
addresses the priorities set forth in the IOM reports and serves as a guide for CDC
and its partners to combat emerging microbial threats to health. Laboratory-based
surveillance, better communication networks, and improvements in the public health
infrastructure are the cornerstones of the strategy. Emerging Infectious Diseases, a
new periodical produced by CDC, will serve as a forum for exchange of information
about incipient trends in infectious diseases, analysis of factors contributing to disease
emergence, and development and implementation of prevention measures.
“Nothing in the world of living things is permanently fixed.”

Hans Zinnser—Rats, Lice and History, 1935
Early History of Infectious Diseases water were additional sources of infectious microor-
Infectious diseases have plagued humans since ganisms.
the dawn of civilization (1-5). The history of these Dissemination of infectious diseases intensified
diseases provides a valuable perspective for evalu- as civilizations progressed. Caravans of traders car-
ating current trends. Humans are presumed to have ried new pathogens to unsuspecting and susceptible
originated in tropical climates and to have been populations. Explorers and later conquering armies
affected by the same parasitic diseases as other brought infectious microorganisms to new conti-
primates in these areas. As available supplies of nents. Stowaway rats and other vermin in the holds
game diminished, early hunters migrated into tem- of ships traveled down the moorings when the ships
perate zones which were free of tropical parasites. docked, bringing fleas, lice, and deadly pathogens to
Historians speculate that humans were relatively a new world. Sporadic epidemics of plague, small-
safe from infectious diseases during that period. pox, typhus, and measles ravaged cities, decimated
Later, however, as agriculture began to provide a armies, and altered the course of history.
substantial portion of the human diet, populations
stabilized and grew. Eventually, populations
reached a size that would support persistent person- 19th Century Discoveries Lead to
to-person spread of infectious microorganisms. With Infectious Disease Prevention and Control
this newly established mode of transmission, infec- Control of many infectious diseases became pos-
tious diseases soon became widespread. The exact sible with the pioneering work of Robert Koch and
origins of many infectious agents remain obscure, Louis Pasteur and the introduction of the germ
but with the advent of large populations, humans theory of disease. With bacteriologic cultivation
eventually became the established reservoir of many techniques came the first isolation and identifica-
agents. Infected animals and contaminated food and tion of etiologic agents; virus cultivation and identi-
fication became available some decades later.
Reservoirs of microorganisms and their life cycles
Address for correspondence: David Satcher, Centers for were identified; the epidemiology and natural his-
Disease Control and Prevention, 1600 Clifton Road, tory of many infectious diseases were described, and
Atlanta, GA 30333, USA, fax 404-639-3039, e-mail successful control measures were initiated. Water
eideditor@cidod1.em.cdc.gov.
Vol. 1, No. 1 — January-March 1995 1 Emerging Infectious Diseases

Perspectives
treatment, vector control, and rodent reduction pro- 600,000 cases of pneumonia occur in the United
grams followed. By the beginning of the 20th cen- States each year and cause 25,000 to 50,000 deaths
tury, the principles of vaccination, established (11). More than 10,000 cases of diphtheria have
empirically by Edward Jenner more than 100 years occurred in Russia since 1993 because of inadequate
earlier, began to be realized in earnest. Antibiotics levels of immunization (12). Despite a century of
were discovered, and disinfectants were developed. scientific progress, infectious diseases still cause
Collectively, these control measures dramatically enormous human suffering, deplete scarce re-
decreased the incidence and prevalence of many sources, impede social and economic development,
infectious diseases and their fatality rates. The early and contribute to global instability. The potential for
part of this century is appropriately regarded as a even greater dissemination looms as a continuous
golden age in public health. threat.
Recent outbreaks underscore the potential for the
sudden appearance of infectious diseases in cur-
New and Reemerging Infectious rently unaffected populations. In the United States,
Diseases—A Contemporary Problem contamination of the municipal water supply in Mil-
Compared with earlier generations, we possess waukee, Wisconsin, in 1993 resulted in an outbreak
an enormous scientific base, and the rate of acquisi- of cryptosporidiosis that affected an estimated
tion of new information about infectious diseases is 400,000 people; approximately 4,400 persons re-
at a historic high. Moreover, thanks in large measure quired hospitalization (13). In the 1990s, epidemic
to effective childhood immunization programs, in- cholera reappeared in the Americas, after being ab-
cluding the President’s Childhood Immunization In- sent for nearly a century; from 1991 through June
itiative, many infectious diseases are under control, of 1994 more than one million cases and nearly
particularly in the industrialized world. The elimi- 10,000 deaths were reported (14). During the 1980s,
nation of smallpox in 1977 stands as a towering tuberculosis reemerged in the United States after
achievement in the fight against infectious diseases. decades of decline, and drug-resistant strains have
However, many infectious diseases have persisted made its control more difficult (15,16). The increas-
and have displayed a remarkable ability to re- ing prevalence of antibiotic-resistant strains of
emerge after lengthy periods of stability. Therefore, gonococci, pneumococci, enterococci, and staphylo-
we must be ever mindful of the cyclical nature of cocci portend of other serious treatment and control
disease trends. failures. Many other examples of emerging infec-
A careful review of infectious disease trends tions could be given (17,18).
shows a fragile equilibrium between humans and New infectious diseases, often with unknown
infectious microorganisms. Infectious diseases are long-term public health impact, continue to be iden-
still broadly endemic and maintain a large reservoir tified. Table 1 lists major diseases or etiologic agents
of agents that have the potential for rapid and wide- identified just within the last 20 years (19-41). New
spread dissemination. Infectious diseases remain agents are regularly added to the list, particularly
the leading cause of death worldwide, even as the with the availability of nucleic acid amplification
International Code of Diseases places many infec- techniques for detecting and identifying otherwise
tious diseases in other categories. For example, men- noncultivable microorganisms (40, 42).
ingitis and cirrhosis are classified as diseases of the In some cases, etiologic agents have been identi-
nervous system and liver, respectively, and only 17% fied as the causes of known diseases or syndromes
of deaths attributable to infections are actually in- (e.g., rotavirus, parvovirus, human T-cell lympho-
cluded in the codes for parasitic and infectious dis- tropic viruses I and II (HTLV I/II), and human her-
eases (6). In the United States, each year, pesvirus type 6, (HHV-6); in other cases, diseases
approximately 25% of physician visits are attribut- became better recognized or defined (e.g., Legion-
able to infectious diseases, with direct and indirect naires’ disease, Lyme disease, human ehrlichiosis).
costs, including those for human immunodeficiency Still others are entirely new: with some parallels to
virus (HIV) infection and related illnesses, esti- medieval times, a previously unknown and deadly
mated at more than $120 billion (7). disease, acquired immunodeficiency syndrome
Persons living in tropical climates are still as (AIDS), originated from uncertain sources in one
vulnerable to infectious diseases as their early an- part of the world and became globally disseminated;
cestors were. Each year more than one million chil- this time the disease spread at a rate that would
dren die of malaria in sub-Saharan Africa alone (8); have been unthinkable in medieval times. Clearly,
worldwide, approximately 200 million people have the complete history of infectious diseases remains
schistosomiasis (9), and each year 35-60 million peo- to be written.
ple contract dengue (10). Moreover, infectious dis-
eases and their attendant problems are not confined
to tropical climates. For example, an estimated
Emerging Infectious Diseases 2 Vol. 1, No. 1 — January-March 1995

Perspectives
Getting Ahead of the Curve surveillance to be effective it must be specific. Con-

Recent disquieting infectious disease trends have sider, for example, surveillance of viral hepatitis.
not gone unrecognized, and their similarity to earlier Only after the various agents of viral hepatitis were
disease trends with immense global consequences identified and specific laboratory testing became
has not gone unnoticed. Primary responsibility for available was it possible to explain trends in disease
addressing new and reemerging infectious diseases prevalence and establish the epidemiologic princi-
rests squarely with the custodians of public health. ples underlying the different modes of transmission.
Indeed, the fundamental maxim of public health Specific laboratory testing is also the basis of screen-
must guide current prevention programs: the health ing programs that ensure the safety of the blood
of the individual is best ensured by maintaining or supply against hepatitis B and hepatitis C. Agent-
improving the health of the entire community. Core specific surveillance is a critical component of many
functions necessary to ensure the health of the pub- immunization programs. Vaccines to Haemophilus
lic were defined in the National Academy of Science’s influenzae type b, (Hib), for example, were developed
Institute of Medicine (IOM) report on The Future of in response to laboratory-based surveillance that
Public Health (43): identified Hib as a major cause of invasive disease
• Assessment of health status, risks, and services in children. The effectiveness of the Hib vaccination
campaign in the United States has been dramatic
• Development of health policy
(Figure 1). Similar approaches will ensure appropri-
• Assurance of quality health services ate formulation of other developmental vaccines.
Surveillance (assessment) is the sine qua non of Monitoring antibiotic resistance is yet another im-
infectious disease prevention programs; however, for portant example of the value of laboratory-based
Table 1. Major Etiologic Agents, Infectious Diseases Identified Since 1973*

Year Agent Disease Reference
1973 Rotavirus Major cause of infantile diarrhea worldwide 19
1975 Parvovirus B19 Fifth disease; Aplastic crisis in chronic hemolytic 20
anemia
1976 Cryptosporidium parvum Acute enterocolitis 21
1977 Ebola virus Ebola hemorrhagic fever 22
1977 Legionella pneumophila Legionnaires’ disease 23
1977 Hantaan virus Hemorrhagic fever with renal syndrome (HFRS) 24
1977 Campylobacter sp. Enteric pathogens distributed globally 25
1980 Human T-cell lymphotropic T-cell lymphoma—leukemia 26
virus-I (HTLV I)
1981 Staphylococcus toxin Toxic shock syndrome associated with tampon use 27
1982 Escherichia coli O157:H7 Hemorrhagic colitis; hemolytic uremic syndrome 28
1982 HTLV II Hairy cell leukemia 29
1982 Borrelia burgdorferi Lyme disease 30
1983 Human immunodeficiency virus Acquired immunodeficiency syndrome (AIDS) 31
(HIV)
1983 Helicobacter pylori Gastric ulcers 32
1988 Human herpesvirus-6 (HHV-6) Roseola subitum 33
1989 Ehrlichia chaffeensis Human ehrlichiosis 34
1989 Hepatitis C Parenterally transmitted non-A, non-B hepatitis 35
1991 Guanarito virus Venezuelan hemorrhagic fever 36
1992 Vibrio cholerae O139 New strain associated with epidemic cholera 37
1992 Bartonella (= Rochalimaea) Cat-scratch disease; bacillary angiomatosis 38, 39
henselae
1993 Hantavirus isolates Hantavirus pulmonary syndrome 40
1994 Sabiá virus Brazilian hemorrhagic fever 41
* Compiled by CDC staff. Dates of discovery are assigned on the basis of the year the isolation or identification of etiologic agents
was reported.

Perspectives
surveillance. Within this context, current discover- risk for drug-resistant tuberculosis in the commu-
ies of etiologic agents and diseases (Table 1) are nity.
reasons for optimism. The potential for improve- Microbiologists, infectious disease specialists,
ments in assessment and prevention of these and and other basic and clinical scientists must collabo-
other newly discovered diseases is reminiscent of the rate with behavioral scientists in an interdiscipli-
watershed years of Koch and Pasteur. nary effort to prevent and control emerging
We cannot overstate the role of behavioral science infections.
in our effort to “get ahead of the curve” with emerg- The Future of Public Health emphasizes the rela-
ing infections. Having the science or laboratory tech- tionship between a sound public health infrastruc-
nology to control infectious diseases is not enough, ture and infectious disease prevention programs.
unless we can influence people to behave in ways Infrastructure improvements must become a na-
that minimize the transmission of infections and tional priority: certainly they are among CDC’s top
maximize the efforts of medical interventions. For priorities. Improvements in infectious disease sur-
example, even though HIV/AIDS does not have a veillance are particularly needed (44). Enriching the
vaccine or cure, it is almost entirely preventable. For capacity to respond to urgent threats to health and
many people, however, reducing the risk for HIV developing nationwide prevention strategies are
infection and AIDS requires important changes in also CDC priorities. To combat new and reemerging
lifestyle or behavior. We must use our knowledge of infectious diseases, CDC, in collaboration with other
human behavior to help people make lifestyle federal agencies, state and local health depart-
changes and prevent disease. ments, academic institutions, professional societies,
Another illustration of the need to use behavior international organizations, and experts in public
science is the problem of antibiotic resistance. To a health infectious diseases and medical microbiology
great extent, this problem is related to the behavior developed a plan—Addressing Emerging Infectious
of both physicians and patients. Physicians continue Disease Threats: A Prevention Strategy for the
to use antibiotics inappropriately, and patients con- United States (7). The plan has four major goals:
tinue to demand antibiotic treatment when it is not • Surveillance and Response—detect, promptly
indicated, for example, for the common cold. As soci- investigate, and monitor emerging pathogens,
ety changes and institutions such as day care centers the diseases they cause, and the factors influenc-
and prisons become more crowded, the spread of ing their emergence
infectious diseases is exacerbated. For homeless and • Applied research—integrate laboratory sci-
drug-dependent populations, completing a 6- to 9- ence and epidemiology to optimize public health
month course of therapy for tuberculosis is difficult, practice
and the failure to complete the therapy increases the • Prevention and control—enhance communi-
cation of public health information about emerg-
ing diseases and ensure prompt implementation
of prevention strategies
• Infrastructure—strengthen local, state, and
federal public health infrastructures to support
surveillance and implement prevention and con-
trol programs
CDC’s plan provides a framework for the agency
to work collaboratively with its many partners to
identify and reverse worrisome trends in infectious
diseases.
The need for implementing CDC’s plan is urgent,
given the extremely dynamic nature of disease
Figure 1. Race-adjusted incidence rate* of trends and the complexity of factors contributing to
Haemophilus influenzae type b and non-type b disease disease emergence; these were outlined in detail in
detected through laboratory-based surveillance†
the 1992 IOM report—Emerging Infections: Micro-
among children aged <5 years — United States,
1989–1993 bial Threats to Health in the United States (45) and
are discussed in a companion article by Stephen S.
* Per 100,000 children aged <5 years.
† Morse, Ph.D., in this issue. The IOM report concludes
The surveillance area population was 10.4 million in four
states (three counties in the San Francisco Bay Area, eight that infectious diseases must be viewed as but one
counties in metropolitan Atlanta, four counties in Tennessee, component of a dynamic and complex global ecology,
and the state of Oklahoma. which is shaped and buffeted by technologic, socie-
Source: CDC. Progress toward elimination of Haemophilus tal, economic, environmental, and demographic
influenzae type b disease among infants and children
— United States, MMWR 1994;43:144-8.

Perspectives
changes, not to mention microbial change and adap- 5. McNeill WH. Plagues and peoples. Garden City, New
tation. York: Anchor Press/Doubleday, 1976.
Clearly, broader coalitions are needed, and com- 6. Bennett JV, Holmberg SD, Rogers MF, Solomon SL.
Infectious and parasitic diseases. In: Amler RW, Dull
munication must improve if we are to “get ahead of
HB, editors. Closing the gap: the burden of unneces-
the curve.” This new periodical is part of the overall sary illness. New York: Oxford University Press, 1987.
strategy to draw worldwide attention to emerging 7. Centers for Disease Control and Prevention. Address-
infections and improve communication. Given the ing emerging infectious disease threats: a prevention
multiplicity of factors contributing to disease emer- strategy for the United States. Atlanta, Georgia: U.S.
gence, Emerging Infectious Diseases (EID) will pre- Department of Health and Human Services, Public
sent relevant concepts from professionals in Health Service, 1994.
multiple disciplines and disseminate information 8. World Health Organization. Severe and complicated
about emerging infectious diseases in order to de- malaria. Trans R Soc Trop Med Hyg 1990; 84:Supp
velop and apply ecologically acceptable interven- 2:1-65.
tions that will benefit humankind. Prevention and 9. World Health Organization. Tropical disease re-
search: progress 1991-92—Eleventh Programme Re-
control of new and emerging infectious diseases de- port of the UNDP/World Bank, WHO Special
pend on the participation of scientists and other Programme for Research and Training in Tropical
professionals in the public and private sectors. Diseases (TDR). Geneva: World Health Organization,
CDC will make EID available by print and elec- 1993.
tronically to facilitate rapid communication. We 10. Gubler DJ. Vector-borne diseases. In: Encyclopedia of
hope that in the process EID will promote the ex- the environment. New York: Houghton Mifflin Co.,
change of infectious disease information through 1994.
global electronic networks and bulletin boards. 11. Marston BJ, Plouffe JF, Breiman RF, et al. Prelimi-
nary findings of a community-based pneumonia inci-
Although there are many similarities between dence study. In: Barbaree JM, Breiman RF, Dufour AP,
our vulnerability to infectious diseases and that of editors. Legionella: current status and emerging per-
our ancestors, there is one distinct difference: we spectives. Washington, D.C.; American Society for Mi-
have the benefit of extensive scientific knowledge. crobiology, 1993.
Ultimately, our success in combatting infectious dis- 12. Centers for Disease Control and Prevention. Diphthe-
eases will depend on how well we use available ria outbreak—Russian Federation, 1990-1993.
information. A recent report by the Carnegie Com- MMWR 1993; 42:840-1, 847.
mission “Science, Technology, and Government for a 13. MacKenzie WR, Hoxie NJ, Proctor ME, et al. A mas-
Changing World,” provides valuable insight in this sive outbreak in Milwaukee of Cryptosporidium infec-
tion transmitted through the public water supply. N
regard (46). Commenting on the Earth Summit in
Engl J Med 1994; 331:161-7.
Rio de Janeiro in 1992, the report emphasizes the
14. Organizacion Panamericana de la Salud. El colera en
need to shift from the “manifestations of environ- las Americas. Informe No. 10; Junio 1994.
mental changes in the air, land, water, and plant and 15. Centers for Disease Control and Prevention. Ex-
animal kingdoms to the causes of those changes.” panded tuberculosis surveillance and tuberculosis
Indeed, the advice of that report challenges us all— morbidity—United States, 1993. MMWR 1994;
"our capacity to generate, integrate, disseminate, 43:361-6.
and apply knowledge will determine the human 16. Centers for Disease Control and Prevention. Multi-
prospect in the 21st century." drug-resistant tuberculosis in a hospital—Jersey City,
New Jersey, 1990-1992. MMWR 1994; 43:417-9.
Dr. Satcher is director of the Centers for Disease 17. Murphy FA. New, emerging, and reemerging infec-
tious diseases. Adv Virus Res 1994; 43: 1-52.
Control and Prevention, Atlanta, Georgia. He was
18. Morse SS, editor. Emerging viruses. New York: Oxford
president of Meharry Medical College, Nashville,
University Press, 1993.
Tennessee, from 1982 to 1993 and is a former faculty
19. Bishop RF, Davidson GP, Holmes IH, Ruck BJ. Virus
member of the King-Drew Medical Center and the particles in epithelial cells of duodenal mucosa from
UCLA School of Medicine, Los Angeles, California. children with acute non-bacterial gastroenteritis.
Lancet 1973; 2:1281-3.
20. Cossart YE, Field AM, Cant B, Widdows D. Par-
References vovirus-like particles in human sera. Lancet 1975;
1. Zinsser H. Rats, lice and history. Boston: Little, 1:72-3.
Brown, and Company, 1935. 21. Nime FA, Burek JD, Page DL, Holscher MA, Yardley
2. Hopkins DR. Princes and peasants: smallpox in his- JH. Acute enterocolitis in a human being infected with
tory. Chicago: University of Chicago Press, 1983. the protozoan Cryptosporidium. Gastroenterology
3. Bollet AJ. Plagues and poxes. New York: Demos Pub- 1976; 70: 592-8.
lications, 1987. 22. Johnson KM, Webb PA, Lange JV, Murphy FA. Isola-
4. Burnet M, White DO. Natural history of infectious tion and partial characterization of a new virus caus-
disease. London: Cambridge University Press, 1972. ing acute haemorrhagic fever in Zaire. Lancet 1977;
1:569-71.

Perspectives
23. McDade JE, Shepard CC, Fraser DW, Tsai TR, Redus 36. Salas R, de Manzione N, Tesh RB, Rico-Hesse R,
MA, Dowdle WR, Laboratory Investigation Team. Le- Shope RE, Betancourt A, Goday O, Bruzual R,
gionnaires’ disease. 2: Isolation of a bacterium and Pacheco ME, Ramos B, Taibo ME, Tamayo JG, Jaimes
demonstration of its role in other respiratory disease. E, Vasquez C, Araoz, F, Querales J. Venezuelan hem-
N Engl J Med 1977; 297:1197-1203. orrhagic fever. Lancet 1991; 338:1033-6.
24. Lee HW, Lee PW, Johnson KM. Isolation of the etio- 37. World Health Organization. Epidemic diarrhea due to
logic agent of Korean hemorrhagic fever. J Infect Dis Vibrio cholerae non-01. Wkly Epidemiol Rec 1993;
1978; 137:298-308. 68:141-2.
25. Skirrow MB. Campylobacter enteritis: a “new” dis- 38. Regnery RL, Anderson BE, Clarridge JE III, Ro-
ease. Br Med J 1977; 2:9-11. dríguez-Barradas MC, Jones DC, Carr JH. Charac-
26. Poiesz BJ, Ruscetti FW, Gazdar AF, Bunn PA, Minna terization of a novel Rochalimaea species, R. henselae
JD, Gallo RC. Detection and isolation of type C sp. nov., isolated from blood of a febrile, human immu-
retrovirus particles from fresh and cultured lympho- nodeficiency virus-positive patient. J Clin Microbiol
cytes of a patient with cutaneous T-cell lymphoma. 1992; 30:265-74.
Proc Natl Acad Sci 1980; 77: 7415-9. 39. Welch DF, Pickett DA, Slater LN, Steigerwalt AG,
27. Schlievert PM., Shands KN, Gan BB, Schmid GP, Brenner DJ. Rochalimaea henselae sp. nov., a cause of
Nishimura RD. Identification and characterization of septicemia, bacillary angiomatosis, and parenchymal
an exotoxin from Staphylococcus aureus associated bacillary peliosis. J Clin Microbiol 1992; 30:275-80.
with toxic shock syndrome. J Infect Dis 1981; 143:509- 40. Nichol ST, Spiropoulou CF, Morzunov S, Rollin PE,
16. Ksiazek TG, Feldmann H, Sanchez A, Childs J, Zaki
28. Riley LW, Remis RS, Helgerson SD, et al. Hemor- S, Peters, CJ. Genetic identification of a hantavirus
rhagic colitis associated with a rare Escherichia coli associated with an outbreak of acute respiratory ill-
serotype. N Engl J Med 1983; 308:681-5. ness. Science 1993; 262:914-7.
29. Kalyanaraman S, Sarangadharan MG, Poiesz B, 41. Lisieux T, Coimbra M, Nassar ES, Burattini MN, de
Ruscetti FW, Gallo RC. Immunological properties of a Souza LT, Ferreira I, Rocco IM, daRosa AP, Vascon-
type C retrovirus isolated from cultured human T- celos PF, Pinheiro FP, et al. New arenavirus isolated
lymphoma cells and comparison to other mammalian in Brazil. Lancet 1994; 343:391-2.
retroviruses, J Virol 1981; 38:906-15. 42. Relman DA, Falkow S, LeBoit PE, Perkocha LA, Min
30. Burgdorfer W, Barbour AG, Hayes SF, Benach JL, K-W, Welch DF, Slater LN. The organism causing
Grunwaldt E, Davis JP. Lyme disease-a tick-borne bacillary angiomatosis, peliosis hepatitis, and fever
spirochetosis?. Science 1982; 216:1317-9. and bacteremia in immunocompromised patients. N
31. Barré-Sinoussi F, Chermann JC, Rey F, et al. Isolation Engl J Med 1991; 324:1514.
of a T-lymphotropic retrovirus from a patient at risk 43. Institute of Medicine. The future of public health.
for acquired immune deficiency syndrome. Science Washington, D.C.: National Academy Press, 1988.
1983; 220:868-71. 44. Berkelman RL, Bryan RT, Osterholm MT, LeDuc JW,
32. Marshall B. Comment in: Lancet 1983; 1:1273-5. Hughes JM. Infectious disease surveillance: a crum-
33. Yamanishi K, Okuno T, Shiraki K, Takahashi M, bling foundation. Science 1994; 264: 368-70.
Kondo T, Asano Y, Kurata T. Identification of human 45. Institute of Medicine. Emerging infections: microbial
herpesvirus-6 as a causal agent for exanthem subi- threats to health in the United States. Washington,
tum. Lancet 1988; 1:1065-7. D.C.: National Academy Press, 1992.
34. Dawson JE, Anderson BE, Fishbein DB, Sanchez JL, 46. Malone TF. The institutions of science and the global
Goldsmith CS, Wilson KH, Duntley CW. Isolation and prospect: the case of environment. In: Science, tech-
characterization of an Ehrlichia sp. from a patient nology, and government for a changing world: the
diagnosed with human ehrlichiosis. J Clin Microbiol concluding report of the Carnegie Commission on
1991; 29:2741-5. Science, Technology, and Government. New York:
35. Choo QL, Kuo G, Weiner AJ, Overby LR, Bradley DW, Carnegie Commission, 1993.
Houghton M. Isolation of a cDNA clone derived from
a blood-borne non-A, non-B viral hepatitis genome.
Science 1989; 244:359-61.

Perspectives
Factors in the Emergence of

Infectious Diseases
Stephen S. Morse, Ph.D.
The Rockefeller University, New York, NewYork, USA
“Emerging” infectious diseases can be defined as infections that have newly

appeared in a population or have existed but are rapidly increasing in incidence or
geographic range. Among recent examples are HIV/AIDS, hantavirus pulmonary
syndrome, Lyme disease, and hemolytic uremic syndrome (a foodborne infection
caused by certain strains of Escherichia coli). Specific factors precipitating disease
emergence can be identified in virtually all cases. These include ecological, environ-
mental, or demographic factors that place people at increased contact with a pre-
viously unfamiliar microbe or its natural host or promote dissemination. These
factors are increasing in prevalence; this increase, together with the ongoing evolution
of viral and microbial variants and selection for drug resistance, suggests that
infections will continue to emerge and probably increase and emphasizes the urgent
need for effective surveillance and control. Dr. David Satcher’s article and this
overview inaugurate “Perspectives,” a regular section in this journal intended to
present and develop unifying concepts and strategies for considering emerging
infections and their underlying factors. The editors welcome, as contributions to the
Perspectives section, overviews, syntheses, and case studies that shed light on how
and why infections emerge, and how they may be anticipated and prevented.
Infectious diseases emerging throughout history causes for a number of infections that have emerged
have included some of the most feared plagues of the recently. I have suggested that infectious disease
past. New infections continue to emerge today, while emergence can be viewed operationally as a two-step
many of the old plagues are with us still. These are process: 1) Introduction of the agent into a new host
global problems (William Foege, former CDC direc- population (whether the pathogen originated in the
tor now at the Carter Center, terms them “global environment, possibly in another species, or as a
infectious disease threats”). As demonstrated by in- variant of an existing human infection), followed by
fluenza epidemics, under suitable circumstances, a 2) establishment and further dissemination within
new infection first appearing anywhere in the world the new host population (“adoption”) (4). Whatever
could traverse entire continents within days or its origin, the infection “emerges” when it reaches a
weeks. new population. Factors that promote one or both of
We can define as “emerging” infections that have these steps will, therefore, tend to precipitate dis-
newly appeared in the population, or have existed ease emergence. Most emerging infections, and even
but are rapidly increasing in incidence or geographic antibiotic-resistant strains of common bacterial
range (1,2). Recent examples of emerging diseases pathogens, usually originate in one geographic loca-
in various parts of the world include HIV/AIDS; tion and then disseminate to new places (5).
classic cholera in South America and Africa; cholera Regarding the introduction step, the numerous
due to Vibrio cholerae O139; Rift Valley fever; han- examples of infections originating as zoonoses (7,8)
tavirus pulmonary syndrome; Lyme disease; and suggest that the “zoonotic pool”—introductions of
hemolytic uremic syndrome, a foodborne infection infections from other species—is an important and
caused by certain strains of Escherichia coli (in the potentially rich source of emerging diseases; peri-
United States, serotype O157:H7). odic discoveries of “new” zoonoses suggest that the
Although these occurrences may appear inexpli- zoonotic pool appears by no means exhausted. Once
cable, rarely if ever do emerging infections appear introduced, an infection might then be disseminated
without reason. Specific factors responsible for dis- through other factors, although rapid course and
ease emergence can be identified in virtually all high mortality combined with low transmissibility
cases studied (2-4). Table 1 summarizes the known are often limiting. However, even if a zoonotic agent
is not able to spread readily from person to person
and establish itself, other factors (e.g., nosocomial
Address for correspondence: Stephen S. Morse, The
Rockefeller University, 1230 York Avenue, Box 120, New
infection) might transmit the infection. Additionally,
York, NY 10021-6399, USA; fax 212-327-7172; e-mail if the reservoir host or vector becomes more widely
morse@rockvax.rockefeller.edu. disseminated, the microbe can appear in new places.

Perspectives
Table 1. Recent examples of emerging infections and probable factors in their emergence
Infection or Agent Factor(s) contributing to emergence
Viral
Argentine, Bolivian hemorrhagic Changes in agriculture favoring rodent host
fever
Bovine spongiform Changes in rendering processes
encephalopathy (cattle)
Dengue, dengue hemorrhagic fever Transportation, travel, and migration; urbanization
Ebola, Marburg Unknown (in Europe and the United States, importation of monkeys)
Hantaviruses Ecological or environmental changes increasing contact with rodent hosts
Hepatitis B, C Transfusions, organ transplants, contaminated hypodermic apparatus,
sexual transmission, vertical spread from infected mother to child
HIV Migration to cities and travel; after introduction, sexual transmission,
vertical spread from infected mother to child, contaminated hypodermic
apparatus (including during intravenous drug use), transfusions, organ
transplants
HTLV Contaminated hypodermic apparatus, other
Influenza (pandemic) Possibly pig-duck agriculture, facilitating reassortment of avian and
mammalian influenza viruses*
Lassa fever Urbanization favoring rodent host, increasing exposure (usually in homes)
Rift Valley fever Dam building, agriculture, irrigation; possibly change in virulence or
pathogenicity of virus
Yellow fever (in “new” areas) Conditions favoring mosquito vector
Bacterial
Brazilian purpuric fever Probably new strain
(Haemophilus influenzae,
biotype aegyptius)
Cholera In recent epidemic in South America, probably introduced from Asia by ship,
with spread facilitated by reduced water chlorination; a new strain (type
O139) from Asia recently disseminated by travel (similarly to past
introductions of classic cholera)
Helicobacter pylori Probably long widespread, now recognized (associated with gastric ulcers,
possibly other gastrointestinal disease)
Hemolytic uremic syndrome Mass food processing technology allowing contamination of meat
(Escherichia coli O157:H7)
Legionella (Legionnaires’ disease) Cooling and plumbing systems (organism grows in biofilms that form on
water storage tanks and in stagnant plumbing)
Lyme borreliosis (Borrelia Reforestation around homes and other conditions favoring tick vector and
burgdorferi) deer (a secondary reservoir host)
Streptococcus, group A (invasive; Uncertain
necrotizing)
Toxic shock syndrome Ultra-absorbency tampons
(Staphylococcus aureus)
Parasitic
Cryptosporidium, other Contaminated surface water, faulty water purification
waterborne pathogens
Malaria (in “new” areas) Travel or migration
Schistosomiasis Dam building
* Reappearances of influenza are due to two distinct mechanisms: Annual or biennial epidemics involving new variants due to
antigenic drift (point mutations, primarily in the gene for the surface protein, hemagglutinin) and pandemic strains, arising
from antigenic shift (genetic reassortment, generally between avian and mammalian influenza strains).

Perspectives
Bubonic plague transmitted by rodent fleas and natural host carrying the virus that was the precur-
ratborne hantavirus infections are examples. sor to HIV-1 were, therefore, instrumental in the
Most emerging infections appear to be caused by introduction of the virus into humans. This probably
pathogens already present in the environment, occurred in a rural area. A plausible scenario is
brought out of obscurity or given a selective advan- suggested by the identification of an HIV-2-infected
tage by changing conditions and afforded an oppor- man in a rural area of Liberia whose virus strain
tunity to infect new host populations (on rare resembled viruses isolated from the sooty mangabey
occasions, a new variant may also evolve and cause monkey (an animal widely hunted for food in rural
a new disease) (2,4). The process by which infectious areas and the putative source of HIV-2) more closely
agents may transfer from animals to humans or than it did strains circulating in the city (11). Such
disseminate from isolated groups into new popula- findings suggest that zoonotic introductions of this
tions can be called “microbial traffic” (3,4). A number sort may occur on occasion in isolated populations
of activities increase microbial traffic and as a result but may well go unnoticed so long as the recipients
promote emergence and epidemics. In some cases, remain isolated. But with increasing movement
including many of the most novel infections, the from rural areas to cities, such isolation is increas-
agents are zoonotic, crossing from their natural ingly rare. After its likely first move from a rural
hosts into the human population; because of the area into a city, HIV-1 spread regionally along high-
many similarities, I include here vector-borne dis- ways, then by long distance routes, including air
eases. In other cases, pathogens already present in travel, to more distant places. This last step was
geographically isolated populations are given an critical for HIV and facilitated today’s global pan-
opportunity to disseminate further. Surprisingly demic. Social changes that allowed the virus to reach
often, disease emergence is caused by human ac- a larger population and to be transmitted despite its
tions, however inadvertently; natural causes, such relatively low natural transmissibility were instru-
as changes in climate, can also at times be respon- mental in the success of the virus in its newfound
sible (6). Although this discussion is confined largely human host. For HIV, the long duration of infectivity
to human disease, similar considerations apply to allowed this normally poorly transmissible virus
emerging pathogens in other species. many opportunities to be transmitted and to take
Table 2 summarizes the underlying factors re- advantage of such factors as human behavior (sex-
sponsible for emergence. Any categorization of the ual transmission, intravenous drug use) and chang-
factors is, of course, somewhat arbitrary but should ing technology (early spread through blood
be representative of the underlying processes that transfusions and blood products) (Table 1).
cause emergence. I have essentially adopted the
categories developed in the Institute of Medicine
Ecological Changes and Agricultural
report on emerging infections (12), with additional
definitions from the CDC emerging infections plan Development
(13). Responsible factors include ecological changes, Ecological changes, including those due to agri-
such as those due to agricultural or economic devel- cultural or economic development, are among the
opment or to anomalies in climate; human demo- most frequently identified factors in emergence.
graphic changes and behavior; travel and commerce; They are especially frequent as factors in outbreaks
technology and industry; microbial adaptation and of previously unrecognized diseases with high case-
change; and breakdown of public health measures. fatality rates, which often turn out to be zoonotic
Each of these will be considered in turn. introductions. Ecological factors usually precipitate
Ecological interactions can be complex, with sev- emergence by placing people in contact with a natu-
eral factors often working together or in sequence. ral reservoir or host for an infection hitherto unfa-
For example, population movement from rural areas miliar but usually already present (often a zoonotic
to cities can spread a once-localized infection. The or arthropod-borne infection), either by increasing
strain on infrastructure in the overcrowded and proximity or, often, also by changing conditions so
rapidly growing cities may disrupt or slow public as to favor an increased population of the microbe or
health measures, perhaps allowing establishment of its natural host (2,4). The emergence of Lyme dis-
the newly introduced infection. Finally, the city may ease in the United States and Europe was probably
also provide a gateway for further dissemination of due largely to reforestation (14), which increased the
the infection. Most successful emerging infections, population of deer and the deer tick, the vector of
including HIV, cholera, and dengue, have followed Lyme disease. The movement of people into these
this route. areas placed a larger population in close proximity
to the vector.
Consider HIV as an example. Although the pre-
cise ancestry of HIV-1 is still uncertain, it appears Agricultural development, one of the most com-
to have had a zoonotic origin (9,10). Ecological fac- mon ways in which people alter and interpose them-
tors that would have allowed human exposure to a selves into the environment, is often a factor

Perspectives
Table 2. Factors in infectious disease emergence*

Factor Examples of specific factors Examples of diseases
Ecological changes (including Agriculture; dams, changes in Schistosomiasis (dams); Rift Valley fever
those due to economic water ecosystems; (dams, irrigation); Argentine hemorrhagic
development and land use) deforestation/reforestation; fever (agriculture); Hantaan (Korean
flood/drought; famine; climate hemorrhagic fever) (agriculture);
changes hantavirus pulmonary syndrome,
southwestern US, 1993 (weather
anomalies)
Human demographics, Societal events: Population Introduction of HIV; spread of dengue; spread
behavior growth and migration of HIV and other sexually transmitted
(movement from rural areas diseases
to cities); war or civil conflict;
urban decay; sexual behavior;
intravenous drug use; use of
high-density facilities
International travel and Worldwide movement of goods “Airport” malaria; dissemination of mosquito
commerce and people; air travel vectors; ratborne hantaviruses;
introduction of cholera into South America;
dissemination of O139 V. cholerae
Technology and industry Globalization of food supplies; Hemolytic uremic syndrome (E. coli
changes in food processing contamination of hamburger meat), bovine
and packaging; organ or spongiform encephalopathy;
tissue transplantation; drugs transfusion-associated hepatitis (hepatitis
causing immunosuppression; B, C), opportunistic infections in
widespread use of antibiotics immunosuppressed patients,
Creutzfeldt-Jakob disease from
contaminated batches of human growth
hormone (medical technology)
Microbial adaptation and Microbial evolution, response to Antibiotic-resistant bacteria, “antigenic drift”
change selection in environment in influenza virus
Breakdown in public health Curtailment or reduction in Resurgence of tuberculosis in the United
measures prevention programs; States; cholera in refugee camps in Africa;
inadequate sanitation and resurgence of diphtheria in the former
vector control measures Soviet Union
* Categories of factors (column 1) adapted from ref. 12, examples of specific factors (column 2) adapted from ref. 13. Categories
are not mutually exclusive; several factors may contribute to emergence of a disease (see Table 1 for additional information).
(Table 2). Hantaan virus, the cause of Korean hem- Perhaps most surprisingly, pandemic influenza
orrhagic fever, causes over 100,000 cases a year in appears to have an agricultural origin, integrated
China and has been known in Asia for centuries. The pig-duck farming in China. Strains causing the fre-
virus is a natural infection of the field mouse Apode- quent annual or biennial epidemics generally result
mus agrarius. The rodent flourishes in rice fields; from mutation (“antigenic drift”), but pandemic in-
people usually contract the disease during the rice fluenza viruses do not generally arise by this proc-
harvest from contact with infected rodents. Junin ess. Instead, gene segments from two influenza
virus, the cause of Argentine hemorrhagic fever, is strains reassort to produce a new virus that can
an unrelated virus with a history remarkably simi- infect humans (16). Evidence amassed by Webster,
lar to that of Hantaan virus. Conversion of grass- Scholtissek, and others, indicates that waterfowl,
land to maize cultivation favored a rodent that was such as ducks, are major reservoirs of influenza and
the natural host for this virus, and human cases that pigs can serve as “mixing vessels” for new
increased in proportion with expansion of maize mammalian influenza strains (16). Pandemic influ-
agriculture (15). Other examples, in addition to enza viruses have generally come from China.
those already known (2,15), are likely to appear as Scholtissek and Naylor suggested that integrated
new areas are placed under cultivation. pig-duck agriculture, an extremely efficient food

Perspectives
production system traditionally practiced in certain Changes in Human Demographics and

parts of China for several centuries, puts these two Behavior
species in contact and provides a natural laboratory Human population movements or upheavals,
for making new influenza recombinants (17). Web- caused by migration or war, are often important
ster has suggested that, with high-intensity agricul- factors in disease emergence. In many parts of the
ture and movement of livestock across borders, world, economic conditions are encouraging the
suitable conditions may now also be found in Europe mass movement of workers from rural areas to cit-
(16). ies. The United Nations has estimated that, largely
Water is also frequently associated with disease as a result of continuing migration, by the year 2025,
emergence. Infections transmitted by mosquitoes or 65% of the world population (also expected to be
other arthropods, which include some of the most larger in absolute numbers), including 61% of the
serious and widespread diseases (18,19), are often population in developing regions, will live in cities
stimulated by expansion of standing water, simply (24). As discussed above for HIV, rural urbanization
because many of the mosquito vectors breed in allows infections arising in isolated rural areas,
water. There are many cases of diseases transmitted which may once have remained obscure and local-
by water-breeding vectors, most involving dams, ized, to reach larger populations. Once in a city, the
water for irrigation, or stored drinking water in newly introduced infection would have the opportu-
cities. (See “Changes in Human Demographics and nity to spread locally among the population and
Behavior” for a discussion of dengue.) The incidence could also spread further along highways and inter-
of Japanese encephalitis, another mosquito-borne urban transport routes and by airplane. HIV has
disease that accounts for almost 30,000 human been, and in Asia is becoming, the best known bene-
cases and approximately 7,000 deaths annually in ficiary of this dynamic, but many other diseases,
Asia, is closely associated with flooding of fields for such as dengue, stand to benefit. The frequency of
rice growing. Outbreaks of Rift Valley fever in some the most severe form, dengue hemorrhagic fever,
parts of Africa have been associated with dam build- which is thought to occur when a person is sequen-
ing as well as with periods of heavy rainfall (19). In tially infected by two types of dengue virus, is in-
the outbreaks of Rift Valley fever in Mauritania in creasing as different dengue viruses have extended
1987, the human cases occurred in villages near their range and now overlap (25). Dengue hemor-
dams on the Senegal River. The same effect has been rhagic fever is now common in some cities in Asia,
documented with other infections that have aquatic where the high prevalence of infection is attributed
hosts, such as schistosomiasis. to the proliferation of open containers needed for
Because humans are important agents of ecologi- water storage (which also provide breeding grounds
cal and environmental change, many of these factors for the mosquito vector) as the population size ex-
are anthropogenic. Of course, this is not always the ceeds the infrastructure (19). In urban environ-
case, and natural environmental changes, such as ments, rain-filled tires or plastic bottles are often
climate or weather anomalies, can have the same breeding grounds of choice for mosquito vectors. The
effect. The outbreak of hantavirus pulmonary syn- resulting mosquito population boom is comple-
drome in the southwestern United States in 1993 is mented by the high human population density in
an example. It is likely that the virus has long been such situations, increasing the chances of stable
present in mouse populations but an unusually mild transmission cycles between infected and suscepti-
and wet winter and spring in that area led to an ble persons. Even in industrialized countries, e.g.,
increased rodent population in the spring and sum- the United States, infections such as tuberculosis
mer and thus to greater opportunities for people to can spread through high-population density set-
come in contact with infected rodents (and, hence, tings (e.g., day care centers or prisons) (12,26-28).
with the virus); it has been suggested that the Human behavior can have important effects on
weather anomaly was due to large-scale climatic disease dissemination. The best known examples
effects (20). The same causes may have been respon- are sexually transmitted diseases, and the ways in
sible for outbreaks of hantaviral disease in Europe which such human behavior as sex or intravenous
at approximately the same time (21,22). With chol- drug use have contributed to the emergence of HIV
era, it has been suggested that certain organisms in are now well known. Other factors responsible for
marine environments are natural reservoirs for disease emergence are influenced by a variety of
cholera vibrios, and that large scale effects on ocean human actions, so human behavior in the broader
currents may cause local increases in the reservoir sense is also very important. Motivating appropri-
organism with consequent flare-ups of cholera (23). ate individual behavior and constructive action,
both locally and in a larger scale, will be essential
for controlling emerging infections. Ironically, as
AIDS prevention efforts have demonstrated, human

Perspectives
behavior remains one of the weakest links in our seminated rapidly along routes of trade and travel,
scientific knowledge. as have antibiotic-resistant bacteria (5,36).
International Travel and Commerce Technology and Industry

The dissemination of HIV through travel has High-volume rapid movement characterizes not
already been mentioned. In the past, an infection only travel, but also other industries in modern
introduced into people in a geographically isolated society. In operations, including food production,
area might, on occasion, be brought to a new place that process or use products of biological origin,
through travel, commerce, or war (8). Trade between modern production methods yield increased effi-
Asia and Europe, perhaps beginning with the silk ciency and reduced costs but can increase the
route and continuing with the Crusades, brought chances of accidental contamination and amplify the
the rat and one of its infections, the bubonic plague, effects of such contamination. The problem is fur-
to Europe. Beginning in the 16th and 17th centuries, ther compounded by globalization, allowing the op-
ships bringing slaves from West Africa to the New portunity to introduce agents from far away. A
World also brought yellow fever and its mosquito pathogen present in some of the raw material may
vector, Aedes aegypti, to the new territories. Simi- find its way into a large batch of final product, as
larly, smallpox escaped its Old World origins to happened with the contamination of hamburger
wreak new havoc in the New World. In the 19th meat by E. coli strains causing hemolytic uremic
century, cholera had similar opportunities to spread syndrome (37). In the United States the implicated
from its probable origin in the Ganges plain to the E. coli strains are serotype O157:H7; additional
Middle East and, from there, to Europe and much of serotypes have been identified in other countries.
the remaining world. Each of these infections had Bovine spongiform encephalopathy (BSE), which
once been localized and took advantage of opportu- emerged in Britain within the last few years, was
nities to be carried to previously unfamiliar parts of likely an interspecies transfer of scrapie from sheep
the world. to cattle (38) that occurred when changes in render-
Similar histories are being repeated today, but ing processes led to incomplete inactivation of scra-
opportunities in recent years have become far richer pie agent in sheep byproducts fed to cattle (39).
and more numerous, reflecting the increasing vol- The concentrating effects that occur with blood
ume, scope, and speed of traffic in an increasingly and tissue products have inadvertently dissemi-
mobile world. Rats have carried hantaviruses virtu- nated infections unrecognized at the time, such as
ally worldwide (29). Aedes albopictus (the Asian HIV and hepatitis B and C. Medical settings are also
tiger mosquito) was introduced into the United at the front line of exposure to new diseases, and a
States, Brazil, and parts of Africa in shipments of number of infections, including many emerging in-
used tires from Asia (30). Since its introduction in fections, have spread nosocomially in health care
1982, this mosquito has established itself in at least settings (Table 2). Among the numerous examples,
18 states of the United States and has acquired local in the outbreaks of Ebola fever in Africa many of the
viruses including Eastern equine encephalomyelitis secondary cases were hospital acquired, most trans-
(31), a cause of serious disease. Another mosquito- mitted to other patients through contaminated hy-
borne disease, malaria, is one of the most frequently podermic apparatus, and some to the health care
imported diseases in non-endemic-disease areas, staff by contact. Transmission of Lassa fever to
and cases of “airport malaria” are occasionally iden- health care workers has also been documented.
tified. On the positive side, advances in diagnostic tech-
A classic bacterial disease, cholera, recently en- nology can also lead to new recognition of agents
tered both South America (for the first time this that are already widespread. When such agents are
century) and Africa. Molecular typing shows the newly recognized, they may at first often be labeled,
South American isolates to be of the current pan- in some cases incorrectly, as emerging infections.
demic strain (32), supporting the suggestion that the Human herpesvirus 6 (HHV-6) was identified only a
organism was introduced in contaminated bilge few years ago, but the virus appears to be extremely
water from an Asian freighter (33). Other evidence widespread (40) and has recently been implicated as
indicates that cholera was only one of many organ- the cause of roseola (exanthem subitum), a very
isms to travel in ballast water; dozens, perhaps common childhood disease (41). Because roseola has
hundreds, of species have been exchanged between been known since at least 1910, HHV-6 is likely to
distant places through this means of transport have been common for decades and probably much
alone. New bacterial strains, such as the recently longer. Another recent example is the bacterium
identified Vibrio cholerae O139, or an epidemic Helicobacter pylori, a probable cause of gastric ul-
strain of Neisseria meningitidis (34,35) (also exam- cers (42) and some cancers (43,44). We have lived
ples of microbial adaptation and change) have dis- with these diseases for a long time without knowing

Perspectives
their cause. Recognition of the agent is often advan- reemerge if there are breakdowns in preventive
tageous, offering new promise of controlling a pre- measures.
viously intractable disease, such as treating gastric Reemerging diseases are those, like cholera, that
ulcers with specific antimicrobial therapy. were once decreasing but are now rapidly increasing
again. These are often conventionally understood
and well recognized public health threats for which
Microbial Adaptation and Change (in most cases) previously active public health meas-
Microbes, like all other living things, are con- ures had been allowed to lapse, a situation that
stantly evolving. The emergence of antibiotic-resis- unfortunately now applies all too often in both de-
tant bacteria as a result of the ubiquity of veloping countries and the inner cities of the indus-
antimicrobials in the environment is an evolution- trialized world. The appearance of reemerging
ary lesson on microbial adaptation, as well as a diseases may, therefore, often be a sign of the break-
demonstration of the power of natural selection. down of public health measures and should be a
Selection for antibiotic-resistant bacteria (5,36) and warning against complacency in the war against
drug-resistant parasites has become frequent, infectious diseases.
driven by the wide and sometimes inappropriate use Cholera, for example, has recently been raging in
of antimicrobial drugs in a variety of applications South America (for the first time in this century) (50)
(27,45,46). Pathogens can also acquire new antibi- and Africa. The rapid spread of cholera in South
otic resistance genes from other, often nonpatho- America may have been abetted by recent reduc-
genic, species in the environment (36), selected or tions in chlorine levels used to treat water supplies
perhaps even driven by the selection pressure of (34). The success of cholera and other enteric dis-
antibiotics. eases is often due to the lack of a reliable water
Many viruses show a high mutation rate and can supply. These problems are more severe in develop-
rapidly evolve to yield new variants (47). A classic ing countries, but are not confined to these areas.
example is influenza (48). Regular annual epidemics The U.S. outbreak of waterborne Cryptosporidium
are caused by “antigenic drift” in a previously circu- infection in Milwaukee, Wisconsin, in the spring of
lating influenza strain. A change in an antigenic site 1993, with over 400,000 estimated cases, was in part
of a surface protein, usually the hemagglutinin (H) due to a nonfunctioning water filtration plant (51);
protein, allows the new variant to reinfect pre- similar deficiencies in water purification have been
viously infected persons because the altered antigen found in other cities in the United States (52).
is not immediately recognized by the immune
system.
On rare occasions, perhaps more often with non- For our Future
viral pathogens than with viruses (49), the evolution In his accompanying article, Dr. David Satcher
of a new variant may result in a new expression of discusses the history of infectious diseases and the
disease. The epidemic of Brazilian purpuric fever in many infections that, from the dawn of history to the
1990, associated with a newly emerged clonal vari- present, have traveled with the caravans and fol-
ant of Hemophilus influenzae, biogroup aegyptius, lowed the invading armies. The history of infectious
may fall into this category. It is possible, but not yet diseases has been a history of microbes on the
clear, that some recently described manifestations march, often in our wake, and of microbes that have
of disease by group A Streptococcus, such as rapidly taken advantage of the rich opportunities offered
invasive infection or necrotizing fasciitis, may also them to thrive, prosper, and spread. And yet the
fall into this category. historical processes that have given rise to the emer-
gence of “new” infections throughout history con-
tinue today with unabated force; in fact, they are
Breakdown of Public Health Measures accelerating, because the conditions of modern life
and Deficiencies in Public Health ensure that the factors responsible for disease emer-
Infrastructure gence are more prevalent than ever before. Speed of
Classical public health and sanitation measures travel and global reach are further borne out by
have long served to minimize dissemination and studies modeling the spread of influenza epidemics
human exposure to many pathogens spread by tra- (53) and HIV (54,55).
ditional routes such as water or preventable by Humans are not powerless, however, against this
immunization or vector control. The pathogens relentless march of microbes. Knowledge of the fac-
themselves often still remain, albeit in reduced tors underlying disease emergence can help focus
numbers, in reservoir hosts or in the environment, resources on the key situations and areas worldwide
or in small pockets of infection and, therefore, are (3,4) and develop more effective prevention strate-
often able to take advantage of the opportunity to gies. If we are to protect ourselves against emerging
diseases, the essential first step is effective global

Perspectives
disease surveillance to give early warning of emerg- 2. Morse SS. Examining the origins of emerging viruses.
ing infections (3,12,13,56). This must be tied to in- In: Morse SS, ed. Emerging viruses. New York: Oxford
centives, such as national development, and University Press, 1993:10-28.
eventually be backed by a system for an appropriate 3. Morse SS. Regulating viral traffic. Issues Sci Technol
1990;7:81-4.
rapid response. World surveillance capabilities are
4. Morse SS. Emerging viruses: defining the rules for
critically deficient (12,56,57). Efforts, such as the viral traffic. Perspect Biol Med 1991;34:387-409.
CDC plan (13), now under way in the United States 5. Soares S, Kristinsson KG, Musser JM, Tomasz A.
and internationally to remedy this situation are the Evidence for the introduction of a multiresistant clone
essential first steps and deserve strong support. of serotype 6B Streptococcus pneumoniae from Spain
Research, both basic and applied, will also be vital. to Iceland in the late 1980s. J Infect Dis 1993;168:158-
63.
6. Rogers DJ, Packer MJ. Vector-borne diseases, models,
This Journal and the “Perspectives” and global change. Lancet 1993;342:1282-4.
Section 7. Fiennes RW. Zoonoses and the origins and ecology of
Early warning of emerging and reemerging infec- human disease. London: Academic Press, 1978.
tions depends on the ability to identify the unusual 8. McNeill WH. Plagues and peoples. New York: Anchor
as early as possible. Information is, therefore, essen- Press/ Doubleday, 1976.
tial. Hence this journal, which is intended as a 9. Myers G, MacInnes K, Korber B. The emergence of
simian/human immunodeficiency viruses. AIDS Res
peer-reviewed forum for the discussion of concepts Hum Retroviruses 1992;8:373-86.
and examples relevant to emerging infectious dis- 10. Allan JS, Short M, Taylor ME, et al. Species-specific
eases and their causes, and to provide a channel for diversity among simian immunodeficiency viruses
field reports and observations on emerging infec- from African green monkeys. J Virol 1991;65:2816-28.
tions. The “Perspectives” section will provide gen- 11. Gao F, Yue L, White AT, et al. Human infection by
eral overviews dealing with factors in disease genetically diverse SIVSM-related HIV-2 in West Af-
emergence, conceptual syntheses of information, ap- rica. Nature 1992;358:495-9.
proaches for studying or predicting emerging infec- 12. Institute of Medicine. Emerging infections: Microbial
tions, and analyses that shed light on how and why threats to health in the United States (Lederberg J,
infections emerge, and how they may be anticipated Shope RE, Oaks SC Jr, eds). Washington, DC: Na-
tional Academy Press, 1992.
and prevented. Submissions for this section are
13. Centers for Disease Control and Prevention. Address-
warmly invited. In coming issues, Perspectives will
ing emerging infectious disease threats: a prevention
deal in greater detail with many of the factors dis- strategy for the United States. Atlanta, Georgia: US
cussed in this overview article, and with ways to Dept of Health and Human Services, Public Health
dissect steps in the emergence process. Discussion Service, 1994.
of technologies that are broadly applicable to the 14. Barbour AG, Fish D. The biological and social phe-
identification or control of emerging diseases are nomenon of Lyme disease. Science 1993;260:1610-6.
also appropriate for this section. Case studies are 15. Johnson KM. Emerging viruses in context: an over-
welcome if they are used to develop broader lessons. view of viral hemorrhagic fevers. In: Morse SS, ed.
Emerging viruses. New York: Oxford University
Press, 1993:46-7.
Acknowledgments 16. Webster RG, Bean WJ, Gorman OT, Chambers TM,
I thank Dr. John La Montagne, NIAID, for helpful Kawaoka Y. Evolution and ecology of influenza A
viruses. Microbiol Rev 1992;56:152-79.
discussions. Supported by NIH grant RR 03121
17. Scholtissek C, Naylor E. Fish farming and influenza
(from CMP), US DHHS. pandemics. Nature 1988;331:215.
18. World Health Organization. Geographical distribu-
Dr. Morse, “Perspectives” section editor of this tion of arthropod-borne diseases and their principal
journal, is assistant professor of virology at The vectors. Gene va : World Health Organization
Rockefeller University, New York, N.Y. He chaired the (WHO/VBC/89.967), 1989:138-48.
NIH Conference on Emerging Viruses (May 1989) and 19. Monath TP. Arthropod-borne viruses. In: Morse SS,
was a member of the committee on Emerging Microbial ed. Emerging viruses. New York: Oxford University
Threats to Health (and chaired its Task Force on Press, 1993.
Viruses), convened by the Institute of Medicine, 20. Levins R, Epstein PR, Wilson ME, Morse SS, Slooff R,
National Academy of Sciences (ref. 12). Eckardt I. Hantavirus disease emerging. Lancet
1993;342:1292.
21. Le Guenno B, Camprasse MA, Guilbaut JC, Lanoux
References P, Hoen B. Hantavirus epidemic in Europe, 1993.
Lancet 1994;343:114-5.
1. Morse SS, Schluederberg A. Emerging viruses: the
evolution of viruses and viral diseases. J Infect Dis 22. Rollin PE, Coudrier D, Sureau P. Hantavirus epi-
1990;162:1-7. demic in Europe, 1993. Lancet 1994;343:115-6.

Perspectives
23. Epstein, PR, Ford TE, Colwell RR. Marine ecosys- 41. Yamanishi K, Okuno T, Shiraki K, et al. Identification
tems. Lancet 1993;342:1216-9. of human herpesvirus-6 as a causal agent for exan-
24. United Nations. World urbanization prospects, 1990. them subitum. Lancet 1988;i:1065-7.
New York: United Nations, 1991. 42. Peterson WL. Helicobacter pylori and peptic ulcer
25. Gubler DJ, Trent DW. Emergence of epidemic den- disease. N Engl J Med 1991;324:1043-8.
gue/dengue hemorrhagic fever as a public health 43. Nomura A, Stemmermann GN, Chyou P-H, Kato I,
problem in the Americas. Infectious Agents and Dis- Perez-Perez GI, Blaser MJ. Helicobacter pylori infec-
ease 1993;26:383-93. tion and gastric carcinoma among Japanese Ameri-
26. Krause RM. The origin of plagues: old and new. Sci- cans in Hawaii. N Engl J Med 1991;325:1132-6.
ence 1992;257:1073-8. 44. Parsonnet J, Friedman GD, Vandersteen DP, et al.
27. Bloom BR, Murray CJL. Tuberculosis: commentary Helicobacter pylori infection and the risk of gastric
on a reemergent killer. Science 1992;257:1055-64. carcinoma. N Engl J Med 1991;325:1127-31.
28. Hoge CW, Reichler MR, Dominguez EA, et al. An 45. Cohen ML. Epidemiology of drug resistance: implica-
epidemic of pneumococcal disease in an overcrowded, tions for a po st-antimicrobial era. Science
ina dequa tely ventilated ja il. N En gl J Med 1992;257:1050-5.
1994;331:643-8. 46. Neu HC. The crisis in antibiotic resistance. Science
29. LeDuc JW, Childs JE, Glass GE. The hantaviruses, 1992;257:1064-72.
etiologic agents of hemorrhagic fever with renal syn- 47. Domingo E, Holland JJ. Mutation rates and rapid
drome: a possible cause of hypertension and chronic evolution of RNA viruses. In: Morse SS, ed. The evo-
renal disease in the United States. Annu Rev Public lutionary biology of viruses. New York: Raven Press,
Health 1992;13:79-98. 1994:161-84.
30. Centers for Disease Control and Prevention. Aedes 48. Kilbourne ED. The molecular epidemiology of influ-
albopictus introduction into continental Africa, 1991. enza. J Infect Dis 1978;127:478-87.
MMWR 1991;40:836-8. 49. Morse SS. Toward an evolutionary biology of viruses.
31. Centers for Disease Control and Prevention. Eastern In: Morse SS, ed. The evolutionary biology of viruses.
equine encephalitis virus associated with Aedes al- New York: Raven Press, 1994:1-28.
bopictus—Florida, 1991. MMWR 1992;41:115, 121. 50. Glass RI, Libel M, Brandling-Bennett AD. Epidemic
32. Wachsmuth IK, Evins GM, Fields PI, et al. The mo- cholera in the Americas. Science 1992;265:1524-5.
lecular epidemiology of cholera in Latin America. J 51. MacKenzie WR, Hoxie NJ, Proctor ME, et al. A mas-
Infect Dis 1993;167:621-6. sive outbreak in Milwaukee of Cryptosporidium infec-
33. Anderson C. Cholera epidemic traced to risk miscal- tion transmitted through the water supply. N Engl J
culation [News]. Nature 1991;354:255. Med 1994;331:161-7.
34. Moore PS. Meningococcal meningitis in sub-Saharan 52. Centers for Disease Control and Prevention. Assess-
Africa: a model for the epidemic process. Clin Infect ment of inadequately filtered public drinking water—
Dis 1992;14:515-25. Washingto n, D.C., December 1993. MMWR
35. Moore PS, Broome CV. Cerebrospinal meningitis epi- 1994;43:661-3.
demics. Sci Am 1994;271(5):38-45. 53. Longini IM Jr, Fine PEM, Thacker SB. Predicting the
36. Davies J. Inactivation of antibiotics and the dissemi- global spread of new infectious agents. Am J
nation of resistance genes. Science 1994;264:375-82. Epidemiol 1986;123:383-91.
37. Centers for Disease Control and Prevention. Update: 54. Flahault A, Valleron AJ. HIV and travel, no rationale
multistate outbreak of Escherichia coli O157:H7 in- for restrictions. Lancet 1990;336:1197-8.
fections from hamburgers—western United States, 55. Flahault A, Valleron AJ. A method for assessing the
1992-1993. MMWR 1993;42:258-63. global spread of HIV-1 infection based on air travel.
38. Morse SS. Looking for a link. Nature 1990;344:297. Mathematical Population Studies 1992;3:161-71.
39. Wilesmith JW, Ryan JBM, Atkinson MJ. Bovine 56. Henderson DA. Surveillance systems and intergov-
spongiform encephalopathy: epidemiological studies ernmental cooperation. In: Morse SS, ed. Emerging
on the origin. Vet Rec 1991;128:199-203. viruses. New York: Oxford University Press,
40. Inoue N, Dambaugh TR, Pellett PE. Molecular biology 1993:283-9.
of human herpesviruses 6A and 6B. Infectious Agents 57. Berkelman RL, Bryan RT, Osterholm MT, LeDuc JW,
and Disease 1993;26:343-60. Hughes JM. Infectious disease surveillance: a crum-
bling foundation. Science 1994;264:368-70.

Synopsis
Unraveling Mysteries Associated with

Cat-Scratch Disease, Bacillary Angiomatosis,
and Related Syndromes
Russell Regnery, Ph.D., and Jordan Tappero, M.D.
National Center for Infectious Diseases,
Centers for Disease Control and Prevention, Atlanta, Georgia, USA
The search for the infectious agents responsible for cat-scratch disease, bacillary
angiomatosis, and related syndromes has a long and often circuitous history. Recog-
nition of the etiologic agents and a new understanding of the fundamental features
of the epidemiology and natural history of modern day Bartonella (formerly Rochali-
maea)-associated diseases culminate a multipartite story that combines clinical
medicine, traditional microbiology, and novel technological approaches to solve a
long-standing enigma.
The quest for the etiologic agent of cat-scratch nopathy in up to 50% of patients; headache, ano-
disease (CSD) has frequently been described as a rexia, weight loss, nausea and vomiting, sore throat,
mystery (1,2). Indeed, the search has many qualities and splenomegaly may develop. In addition, short-
of a mystery novel; the pursuit has spanned several lived, non-specific maculopapular eruptions,
decades and recently taken several unexpected erythema nodosum, figurate erythemas, and throm-
turns. During this period of important discovery, bocytopenic purpura have been observed (8).
major microbial suspects have undergone name Unusual manifestations of CSD, which occur in
changes, novel microbial culprits have been intro- up to 14% of patients, include Perinaud’s oculoglan-
duced, new groups of affected patients have been dular syndrome (6%), encephalopathy (2%), hepatic
recognized, and yet significant questions remain to granulomas (0.3%), osteomyelitis (0.3%), and pulmo-
be answered. Scientific and medical interest has nary disease (0.2%) (4,5,8). In general, these compli-
been high; approximately 900 publications have cations resolve without sequelae. Perinaud’s
dealt with CSD since the first good clinical descrip- oculoglandular syndrome is manifested by conjuncti-
tion of the disease in 1950 (3). val granuloma, periauricular lymphadenopathy, and
nonsuppurative conjunctivitis. Encephalopathy,
manifested as fever and coma that progress to con-
Clinical Features of CSD vulsions, may last for days to weeks; cerebrospinal
Throughout the life of this evolving mystery, the fluid is unremarkable. Optic neuritis with transient
clinical descriptions of “classical” CSD have re- blindness may also occur.
mained remarkably consistent (Dalton MJ, et al. For many years, CSD has been clinically diag-
Rochalimaea antibody; a new era in the diagnosis of nosed when three of the following four criteria are
cat-scratch disease, submitted for publication; 4-6). met in a patient: 1) history of traumatic cat contact;
CSD is typically a benign and self-limited illness 2) positive skin-test response to CSD skin-test anti-
lasting 6 to 12 weeks in the absence of antibiotic gen; 3) characteristic lymph node lesions; and 4)
therapy. Regional lymphadenopathy (axillary, head negative laboratory investigation for unexplained
and neck, inguinal) is the predominant clinical fea- lymphadenopathy (8). Although biopsy confirmation
ture of CSD; affected nodes are often tender and of CSD has been rarely required (especially in lieu
occasionally suppurate (4-7). Between 25% and 60% of a reliable serologic test—see below), a constant
of patients report a primary cutaneous inoculation pathologic hallmark of CSD-affected tissues has
lesion (0.5- to 1-cm papule or pustule) at the site of been granuloma formation. With hematoxylin and
a cat scratch or bite (5,7). The skin lesions typically eosin stains, the primary inoculation lesion of CSD
develop 3 to 10 days after injury and precede the reveals small areas of frank necrosis surrounded by
onset of lymphadenopathy by 1 to 2 weeks. Low- concentric layers of histiocytes, lymphocytes, and
grade fever and malaise accompany lymphade- nucleated giant cells (9). Affected lymph nodes are
characterized by necrotizing granulomas ringed by
Address for correspondence: Russell Regnery, Centers for
Disease Control and Prevention, 1600 Clifton Road, MS lymphocytic infiltrates and multinucleated giant
G13, Atlanta, GA 30333, USA; fax 404-639-1058; e-mail cells.
rur1@ciddvd1.em.cdc.gov.

Synopsis
Enter Afipia felis Ultimately, the relationships between possible

During the past 44 years, a variety of microbial environmental exposures and BA or CSD were sys-
agents, including herpes viruses and bacteria of the tematically investigated. The first case-control
genera Chlamydia and Pasteurella, have been sus- study conducted among patients with BA found
pected as the causes of CSD (3). A major chapter of traumatic contact with a cat (bite or scratch) to be
the CSD saga unfolded with the 1988 announcement significantly associated with BA disease (22). BA
by the Armed Forces Institute of Pathology that a patients were also more likely than controls to have
bacterial agent had been visualized within CSD-in- a household kitten (a cat <1 year of age). A subsequent
volved lymph nodes by using the Warthin-Starry case-control study of CSD patients found that these
silver stain (10), and a novel bacterial agent had patients were more likely than controls to have trau-
been isolated from a CSD patient’s lymph node (11). matic contact with a cat, to own at least one kitten, and
By 1992, this agent was characterized fully, given to have kittens with fleas (7).
the name Afipia felis (Afipia being a latinized acro- Despite the similarities in histochemical staining
nym for the source of the original isolate, the Armed properties and epidemiology, serious reservations
Forces Institute of Pathology, and felis referring to remained concerning a possible link between the
the presumed vertebrate vector of the human infec- causative agents of CSD and BA. The pathologic
tion), and proclaimed the agent of CSD (12). features of classical CSD (granuloma) and BA (pro-
Although A. felis was declared the putative CSD liferative vascular lesions without granuloma) were
bacillus, evidence of convincing patient humoral or distinctly different, and the two diseases seemed to
cellular immune responses to laboratory cultured A. respond differently to antibiotic therapy. Although
felis antigen was not forthcoming. Despite numerous antimicrobial therapy for BA and CSD have not been
attempts, other laboratories were unable to recover systematically evaluated, the majority of BA pa-
additional isolates of A. felis from CSD patients. In tients evaluated responded quickly to single-agent
addition, although the majority of patients with CSD therapy with either erythromycin or doxycycline
reported exposure to a cat(s), no clear link between (14,23), whereas the symptoms and signs of patients
cats and A. felis was demonstrated. with CSD failed to show consistent rapid resolution
following antibiotic therapy (5). In addition, clini-
cians’ first choices of antibiotics for treating BA and
Enter New Syndromes CSD vary (5,6,14,23).
The story of CSD took a significantly divergent
path with the recognition that opportunistic infec-
tions were an important consideration for patients Enter Rochalimaea henselae
infected with human immunodeficiency virus (HIV). A breakthrough occurred when a novel approach
Bacillary angiomatosis (BA), a newly recognized dis- was used to identify possible prokaryotic ribosomal
ease characterized by cutaneous and subcutaneous DNA extracted from BA skin lesions. When prokary-
vascular lesions containing bacillary organisms otic ribosomal gene DNA extracted from BA lesions
visualized by Warthin-Starry silver staining, was and amplified by polymerase chain reaction (PCR)
described predominantly among HIV-infected pa- was compared with sequenced ribosomal genes from
tients; however, bacterial isolates were not made or other organisms, it become apparent that the agent
identified (13-15). Over the ensuing decade, the associated with BA in this study was related to, but
clinical spectrum of BA was expanded to include not necessarily identical to, the agent of trench fever,
patients with single or multiple vascular lesions Rochalimaea quintana (24).
affecting virtually every organ system, including At nearly the same time in Oklahoma, Rochali-
lymph node, bone, brain, liver (peliosis hepatis), and maea-like organisms were being isolated on blood
spleen (14-17). Independently, an unidentified gram- agar from bacteremic patients (18). Independently
negative pathogen was isolated predominantly from in Houston, Texas, fastidious, slow-growing Rochali-
HIV-infected patients with fever and bacteremia, maea-like isolates were recovered on several occa-
however, these patients lacked cutaneous or sions from the blood of an HIV-infected patient with
parenchymal vascular lesions and were not recog- relapsing fever of unknown origin; like the isolates
nized as BA patients (18). from the Oklahoma patients, the Houston isolate
Because silver staining and electron microscopy was recovered from a patient in the absence of BA or
of both BA and CSD tissue sections revealed bacil- CSD lesions (25). The Houston isolate (Houston-1)
lary organisms indistinguishable from one another, was identified as the prototype isolate of a novel
several authors suggested that BA might represent species of Rochalimaea by using traditional as well
disseminated CSD in the immunocompromised host as genotypic methods, including ribosomal RNA
(17,19-21). In addition, several anecdotal reports of gene analysis similar to that used to identify the
BA described a history of cat contact preceding the nucleic acid found in BA patients’ lesions (25). Al-
onset of disease (22). most simultaneously, the group from Oklahoma had

Synopsis
come to a similar conclusion by using DNA related- Data further substantiating the role of R. hense-
ness data (26); most of their isolates also consisted lae in the etiology of CSD soon followed. The newly
of the novel species, R. henselae. The new species developed serologic test was used to help investigate
designation, first officially used to describe the a possible cluster of CSD cases in Connecticut; 38
Houston-1 isolate, was coined in recognition of the (84%) of 45 suspected CSD cases had elevated Ro-
contribution of Diane Hensel, a microbiologist who chalimaea antibody titers, whereas 4 (3.6%) of 112
had isolated several of the initial organisms in Okla- age-matched controls had detectable antibody titers
homa (18,25,26). Subsequently, Koehler et al. iso- (7). In another investigation, serum samples ob-
lated bacilli directly from cutaneous lesions of tained from 600 prospectively evaluated patients
persons with BA (27); surprisingly, either R. henselae with well-characterized CSD (i.e., persons with his-
or R. quintana was isolated from BA lesions from tory of cat scratch, papule at site of inoculation, and
different HIV-infected patients. enlarged regional lymph node) had a 95% correla-
At this juncture, R. henselae infections had been tion with positive Rochalimaea serology.
described predominantly among immunocom- In 1993, R. henselae was isolated directly from the
promised patients with either BA or fever with bac- lymph nodes of two CSD patients and was identified
teremia. The availability of isolates made it possible by genotypic means; both patients had strong sero-
to develop a test for serologic evidence of Rochali- logic responses to Rochalimaea antigen (30). Evi-
maea infection and to refine PCR methods for iden- dence of R. henselae-specific nucleic acid sequences
tification of Rochalimaea organisms in tissues and were found in 21 (84%) of 25 CSD lymph node tissues
other samples. These methods, together with new submitted to CDC for evaluation (31).
techniques for recovering Rochalimaea species iso- Additional supporting evidence for a Rochali-
lates, were crucial to obtaining a more detailed ac- maea as the cause of CSD came from archival
count not only of BA but also of CSD. sources. Skin-test antigen, used rather extensively
in the past to help diagnose examples of CSD (4,8),
consisted of pasteurized exudate collected from sup-
The Cat-scratch Connection: A Synthesis purative CSD lymph nodes. Among a cohort of CSD
A Rochalimaea genus-specific, indirect fluores- patients who were skin-test positive, 52 (93%) of 56
cence antibody (IFA) test using irradiated whole cell had positive IFA antibody titers to the defined Ro-
antigen from the Houston-1 isolate of R. henselae chalimaea reagents (32). Furthermore, various lots
was developed by the Centers for Disease Control of skin-test antigen were shown by PCR analysis to
and Prevention (CDC) to help identify risk factors contain Rochalimaea nucleic acid sequences (33),
for Rochalimaea-associated disease. Several blinded and R. henselae sequences in particular (34). No
serum samples from both HIV-infected BA patients A. felis DNA sequences could be detected by PCR.
and HIV-infected controls residing in San Francisco These data strongly indicated that microbiologically
were sent to CDC for serologic testing. High-titered undefined skin-test reagents, which had been used
antibodies were identified in serum samples from for many years for the diagnosis and clinical charac-
several of the BA patients (28). Similar high-titered terization of CSD, were in fact R. henselae reagents.
antibodies were not detected for any of non-BA con- Collectively, these data supported a Rochalimaea
trol patients with one exception; a serum sample species etiology for both CSD and BA. Despite nu-
from an HIV-infected patient with CSD also demon- merous attempts, recent efforts to implicate A. felis
strated strong serologic reactivity to R. henselae as a cause of either of these two clinical entities have
antigen. repeatedly failed.
Shortly thereafter, single sera collected from pa-
tients with suspected CSD to look for A. felis anti-
bodies were evaluated with the new R. henselae Felis domesticus: A Reservoir for
serologic test; 36 (88%) of 41 sera were positive (29). Rochalimaea henselae
None of the sera had significantly elevated titers to In addition to epidemiologic data, serologic evi-
A. felis antigen. The same set of sera were coded and dence also implicated domestic cats with Rochali-
resubmitted along with sera taken from other well- maea-associated disease. Rochalimaea-specific IFA
characterized bacterial and viral diseases and tested antibodies were demonstrated in 6 (46%) of 13 pet
again in a blinded manner. The IFA test accurately cats not associated with human disease and among
identified sera of case-patients with suspected CSD. 39 (81%) of 48 cats living in households reporting
In addition, 6 (6%) of 107 sera from ostensibly healthy human CSD in Connecticut (7). Microbiologic evi-
persons, obtained from a commercial vendor, showed dence for the domestic cat as a reservoir for R.
antibody by IFA testing (29). These serologic data were henselae soon followed. R. henselae was isolated over
the first laboratory evidence suggesting that R. hense- a 3-week period from the blood of a single cat not
lae was associated with CSD. linked to human illness (35). Investigations by Koe-
hler et al. established the cat as a reservoir for

Synopsis
R. henselae infection (36). R. henselae was established tal laboratories discard their bacteriological plates
as the cause of cutaneous BA in three or four patients before primary isolates of B. henselae would be ex-
with the disease. R. henselae was isolated from the pected to appear (9-40 days). Extreme sensitivity to
blood of all seven asymptomatic pet cats with which a wide variety of antibiotics, at least in vitro, sug-
these four BA patients had prolonged contact. The gests that residual levels of antibiotics in patients’
prevalence of infection among cats in the greater blood or other tissues (such as lymph node biopsy)
San Francisco Bay region was also studied; 25 (41%) might inhibit Bartonella growth during primary iso-
of 61 pet or impounded cats had asymptomatic R. lation attempts in vitro. Selective medium has yet
henselae bacteremia (36). R. henselae was also de- to be developed. Novel genotypic methods were cru-
tected by both direct culture and PCR from several cial for identification of B. henselae; thus, isolates
cat fleas combed from these bacteremic cats (36). may well have been made in the past but remained
The human body louse (Pediculus humanus) was unidentified.
established as a vector for human-to-human trench As mentioned above, it has become apparent that
fever R. quintana transmission during the First in addition to B. henselae, B. quintana can also be
World War (37). Likewise, B. bacilliformis, a closely another significant cause of BA disease, at least
related organism (see below) found in the mountains among immunocompromised patients in San Fran-
of South America, can be transmitted by another cisco (27). Another focus of B. quintana infections
arthropod, the Phlebotomus sand fly (38). The obser- (“urban trench fever”) has been identified among
vation that related microbes are vectored between homeless alcoholics in Seattle (43,44). How common
humans by arthropods adds credence to the pro- are B. quintana infections; are they louseborne and
posed role of arthropod vectors of CSD. Despite sev- vectored strictly between humans, as was believed
eral suggestions that fleas or possibly ticks (7,36,39) during World War I (37)? B. quintana-associated
are associated with R. henselae disease, no experi- disease has no known link with an alternative ver-
mental data exist to clearly demonstrate that ar- tebrate vector (such as cats).
thropods act as direct vectors. Bartonella elizabethae is known only from a sin-
gle isolate from a man surviving endocarditis follow-
ing aortic valve-replacement surgery (45). Is there
Changes in Nomenclature: Rochalimaea further public health significance to this organism?
becomes Bartonella What additional Bartonella species have yet to be
Genotypic evaluation of members of the genus identified and what diseases may they cause?
Rochalimaea has led to the conclusion that members Members of the genus Bartonella are exquisitely
of the genus are closely related to Bartonella bacil- sensitive to antibiotics in vitro (30,46). Why then do
liformis, the agent of Carrión disease, Oroya fever, CSD patients not respond as rapidly and consis-
and verruga peruana (40). Because of historical tently to antibiotic therapy as BA patients do? One
precedence, the genus designation Bartonella is now hypothesis is that immunocompetent patients some-
applied to all species of the old genus Rochalimaea how sequester infectious organisms beyond the
and replaces the Rochalimaea designation (species reach of antibiotics, whereas immunocompromised
names remain unchanged). patients do not. An alternative hypothesis regarding
Physicians and researchers need to exercise care differential antibiotic responsiveness recognizes
in using the term “bartonellosis.” This term has that many of the signs of CSD are immune mediated;
classically been used to describe the frequently fatal antibiotics, even if effective in neutralizing or killing
syndromes caused by B. bacilliformis. To date, bacteria, may not immediately alleviate long-dura-
B. bacilliformis and its associated syndrome (bar- tion immunologic tissue manifestations of antigen
tonellosis) have been identified exclusively in South stimulation. Conversely, in the absence of the immu-
America (38,41). nologic capability to react to bacterial infection by
forming granulomas, as in the case of severely im-
munocompromised persons with BA, antibiotics are
Remaining Questions for Ongoing and generally effective in alleviating the symptoms and
Future Research signs of infection. Does this suggest that possible
Although B. henselae is now regarded as the etio- non-granulomatous manifestations of CSD (for ex-
logic agent of CSD, as well as a cause of BA, endo- ample, neuroretinitis and encephalopathy) should
carditis (42), and fever with bacteremia, many respond well to the appropriate antibiotic therapy?
questions remain unanswered. For example, why Although BA has been described in immunocom-
did it take so long to isolate and identify B. henselae? petent patients (15), the vast majority of BA patients
Part of the answer probably stems from the require- are immunocompromised (14). What are the factors
ments necessary for growth in vitro, including en- explaining why B. henselae and B. quintana induce
riched, non-selective blood agar incubated over a vascular proliferative lesions, such as BA and
prolonged period in a CO2 atmosphere. Most hospi-

Synopsis
parenchymal bacillary peliosis, almost exclusively 7. Zangwill KM, Hamilton DH, Perkins BA, et al.
in severely immunocompromised patients? Epidemiology, risk factors, and evaluation of a new
diagnostic test. N Engl J Med 1993;329:8-13.
What percentage of the relatively large numbers
8. Warwick WJ. The cat-scratch syndrome, many dis-
of undiagnosed febrile disease among HIV-infected
eases or one disease? Prog Med Virol 1967;9:256-301.
persons is in fact due to Bartonella species infec- 9. Johnson WT, Helwig EB. Cat-scratch disease (his-
tions? The answer to this important question may topathologic changes in the skin). Arch Dermatol
help alleviate significant morbidity among HIV-in- 1969;100:148-54.
fected patients. The potential for selection for drug- 10. Wear DJ, Margileth AM, Hadfield TL, Fisher GW,
resistance during long-term antimicrobial therapy Schlagel CJ, King FM. Cat scratch disease: a bacterial
deserves scrutiny. infection. Science 1983;221:1403-5.
Does the 4%-6% of IFA antibody-positive, ostensi- 11. English CK, Wear DJ, Margileth AM, Lissner CR,
bly healthy “control” study populations suggest a Walsh GP. Cat scratch disease: isolation and culture
relatively common undercurrent of undiagnosed, of the bacterial agent. JAMA 1988;259:1347-52.
subclinical Bartonella-associated disease? 12. Brenner DJ, Hollis DG, Moss CW, English CK, et al.
Proposal to Afipia gen. nov., with Afipia felis sp. nov.
Is it possible to immunize cats and thereby inter- (Formerly the Cat Scratch Bacillus), Afipia clevelan-
rupt B. henselae transmission to humans? Prelimi- denis sp. nov. (Formerly the Cleveland Clinic Strain),
nary data suggest that asymptomatic bacteremia in Afipia broomeae sp. nov., and three unnamed genospe-
cats can be successfully treated with antimicrobial cies. J Clin Micro 1991;29:2450-60.
therapy (36). Once cleared of bacteremia, are these 13. Stoler MH, Bonfiglio TA, Steigbigel RT, Pereira M. An
cats routinely susceptible to reinfection? atypical subcutaneous infection associated with ac-
quired immune deficiency syndrome. Am J Clin
Are the complications occasionally associated
Pathol 1983;80:714-8.
with CSD and BA associated with different strains
14. Koehler JE, Tappero JW. AIDS Commentary: bacil-
of Bartonella species or are the variations in clinical lary angiomatosis and bacillary peliosis in patients
presentation strictly functions of dose, route of in- infected with human immunodeficiency virus. Clin
oculation, and immune status? Infect Dis 1993;17:612-24.
And finally, in what role, if any, will A. felis reap- 15. Tappero JW, Koehler JE, Berger TG, Cockerell CJ, Lee
pear as an agent of human disease? Is A. felis respon- T-H, Busch MP, Stites DP, Mohle-Boetani J, Reingold
sible for the relatively small number of cases of AL, LeBoit PE. Bacillary angiomatosis and bacillary
CSD-like lymphadenopathy that have no evidence of splenitis in immunocompetent adults. Ann Intern
Med 1993;118:363-5.
antibody to B. henselae? Or is there another expla-
16. Perkocha LA, Geaghan SM, Yen TS, et al. Clinical and
nation for the originally proposed association be-
pathological features of bacillary peliosis hepatis in
tween A. felis and CSD that has not yet come to light? association with human immunodeficiency virus in-
The new recognition of the importance of Bar- fection. N Engl J Med 1990;323:1581-6.
tonella-associated diseases will continue to spawn a 17. Kemper CA, Lombard CM, Dersinski SC, Tompkins
host of unanswered related questions. Whereas LS. Visceral bacillary epithelioid angiomatosis: possi-
novel subplots will continue to unfold, the new puz- ble manifestations of disseminated cat scratch disease
zles are no longer totally shapeless, and answers to in the immunocompromised host: a report of two
questions of natural history and epidemiology, en- cases. Am J Med 1990;89:216-22.
hanced diagnosis and treatment, and methods for 18. Slater LN, Welch DF, Hensel D, Coody DW. A new
recognized fastidious gram-negative pathogen as a
disease intervention should now begin to unfold
cause of fever and bacteremia. N Engl J Med
rapidly. 1990;323:1587-93.
19. Black JR, Herrington DA, Hadfield TL, Wear DJ,
Margileth AM, Shigekawa B. Life-threatening cat
References scratch disease in an immunocompromised host. Arch
1. Emmons RW. Cat scratch disease: the mystery finally Intern Med 1986;146:394-6.
solved? Ann Intern Med 1984;100:303-4. 20. Koehler JE, LeBoit PE, Egbert BM, Berger TG. Cuta-
2. Goldsmith MF. Has AFIP debugged the cat scratch neous vascular lesions and disseminated cat scratch
mystery? JAMA 1983;250:2745-7. disease in patients with the acquired immunodefi-
3. Emmons RW, Riggs JL, Schachter J. Continuing the ciency syndrome (AIDS) and AIDS-related complex.
search for the etiology of cat scratch disease. J Clin Ann Intern Med 1988;109:449-55.
Micro Biol 1976;4:112-4. 21. LeBoit PE, Berger TG, Egbert BM, Beckstead JH, Yen
4. Carithers HA. Cat scratch disease: An overview based TS, Stoler MH. Epithelioid haemangioma-like vascu-
on a study of 1,200 patients. Am J Dis Child 1985; lar proliferation in AIDS: manifestation of cat-scratch
139:1124-33. disease bacillus infection? Lancet 1988;1:960-3.
5. Margileth AM. Cat scratch disease. Adv Pediatr Infect 22. Tappero JW, Mohle-Boetani J, Koehler JE, Swami-
Dis 1993;8:1-21. nathan B, Berger TG, LeBoit PE, Smith LL, Wenger
6. Karim AA, Cockerell CJ, Petri WA. Cat scratch dis- JD, Pinner RW, Kemper CA, Reingold AL. The
ease, bacillary angiomatosis, and other infections due epidemiology of bacillary angiomatosis and bacillary
to Rochalimaea. N Engl J Med 1994;330:1509-15. peliosis. JAMA 1993;269:770-5.

Synopsis
23. Tappero JW, Koehler JE. Cat scratch disease and bac- 35. Regnery R, Martin M, Olson J. Naturally occurring
illary angiomatosis [letter]. JAMA 1991;266:1938-39. “Rochalimaea henselae” infection in domestic cat.
24. Relman DA, Loutit JS, Schmidt TM, Falkow S, Tomp- Lancet. 1992;340:557-8.
kins LS. The agent of bacillary angiomatosis: an ap- 36. Koehler JE, Glaser CA, Tappero JW. Rochalimaea
proach to the identification of uncultured pathogens. henselae infection: a new zoonosis with the domestic
N Engl J Med 1990;323:1573-80. cat as reservoir. JAMA 1994;271:531-5.
25. Regnery RL, Anderson BE, Clarridge III, JE, Ro- 37. Strong RP (ed.) Trench fever: Report of Commission,
driguez-Barradas MC, Jones DC, Carr JH. Charac- Medical Research Committee, American Red Cross.
terization of a novel Rochalimaea species, R. henselae, Oxford: Oxford University Press, 1918:40-60.
sp. nov., isolated from blood of a febrile, human immu- 38. Schultz MG. A history of bartonellosis (Carrión’s dis-
nodeficiency virus-positive patient. J Clin Microbiol ease). Am J Trop Med Hyg 1968;17:503-15.
1992;30:265-74. 39. Lucey D, Dolan MJ, Moss CW, Garcia M, Hollis DG,
26. Welch DF, Pickett DA, Slater LN, Steigerwalt AG, Wegner S, Morgan G, Almeida R, Leong D, Greisen
Brenner DJ. Rochalimaea henselae, sp. nov., a cause KS, Welch DF, Slater LN. Relapsing illness due to
of septicemia, bacillary angiomatosis, and parenchy- Rochalimaea henselae in immunocompetent hosts:
mal bacillary peliosis. J Clin Microbiol 1992;30:275- Implication for therapy and new epidemiological as-
80. sociations. Clin Infect Dis 1992;14:683-8.
27. Koehler JE, Quinn FD, Berger TG, LeBoit PE, Tap- 40. Brenner DJ, O’Connor SP, Winkler HH, Steigerwalt
pero JW. Isolation of Rochalimaea species from cuta- AG. Proposals to unify the genera Bartonella and
neous and osseous lesions of bacillary angiomatosis. Rochalimaea, with descriptions of Bartonella quin-
N Engl J Med 1992;327:1625-31. tana comb. nov., Bartonella vinsonii comb. nov., Bar-
28. Tappero J, Regnery R, Koehler J, Olson J. Detection tonella henselae comb. nov., and Bartonella
of serologic response to Rochalimaea henselae in pa- elizabethae comb. nov., and to remove the family Bar-
tients with bacillary angiomatosis (BA) by im- tonellaceae from the order Rickettsiales. Int J Syst
munofluorescent antibody (IFA) testing. Program Bacteriol 1993;43:777-86.
Abstr. 32nd Interscience Conference on Antimicrobial 41. Noguchi H, Battistini TS. Etiology of Oroya fever. I.
Agents and Chemotherapy, Anaheim, Calif. Oct 10-14, Cultivation of Bartonella bacilliformis. J Exp Med
1992, Abstr. no. 674. 1926;43:851-64.
29. Regnery RL, Olson JG, Perkins BA, Bibb W. Serological 42. Hadfield TL, Warren R, Kass M, Brun E, Levy C.
response to “Rochalimaea henselae” antigen in sus- Endocarditis caused by Rochalimaea henselae. Hu-
pected cat scratch disease. Lancet 1992;339:1443-5. man Pathol 1993;24:1140-41.
30. Dolan MJ, Wong MT, Regnery RL, Jorgensen JH, 43. Spach DH, Callis KP, Paauw DS, Houze YB, Schoenk-
Garcia M, Peters J, Drehner D. Syndrome of Rochali- necht FD, Welch DF, Rosen H, Brenner DJ. Endo-
maea henselae adenitis suggesting cat scratch dis- carditis caused by Rochalimaea quintana in a patient
ease. Ann Intern Med 1993;118:331-6. infected with human immunodeficiency virus. J Clin
31. Anderson B, Sims K, Regnery R, Robinson L, Schmidt Microbiol 1993;31:692-4.
MJ, Goral S, Hager C, Edwards K. Detection of Ro- 44. Spach DH, Larson AM, Coyle MB, Kanter AS, Welch
chalimaea henselae DNA in specimens from cat DF, Stamm AM. Unanticipated Rochalimaea quin-
scratch disease patients by PCR. J Clin Microbiol tana bacteremia in patients with chronic alcoholism.
1994;32:942-8. [Late Breaker Abstracts]. In: Program Supplement of
32. Szelc Kelly C, Edwards KM, Perez-Perez G, Regnery the 33rd Interscience Conference on Antimicrobial
RL, Perkins BA. A new controversy in the etiology of Agents and Chemotherapy. Washington, D.C.: Ameri-
cat scratch disease: Afipia felis or Rochalimaea hense- can Society for Microbiology, 1993.
lae? Program Abstr. 32nd Interscience Conference on 45. Daly JS, Worthington MG, Brenner DJ, Moss CW,
Antimicrobial Agents and Chemotherapy, Anaheim, Hollis DG, Weyant RS, Steigerwalt AG, Weaver RE,
Calif., Oct 10-14, 1992. Abstr. no. 1565. Daneshvar MI, O’Connor SP. Rochalimaea elizabe-
33. Perkins BA, Swaminathan B, Jackson LA, Brenner thae sp. nov. isolated from a patient with endocarditis.
DJ, Wenger JD, Regnery RL. Case 22-1992—patho- J Clin Microbiol 1993;31:872-81.
genesis of cat scratch disease [letter]. N Engl J Med 46. Myers WF, Grossman DM, Wisseman CL. Antibiotic
1992;327:1599-600. suseptability patterns in Rochalimaea quintana, the
34. Anderson B, Kelley C, Threlkel R, Edwards K. Detec- agent of trench fever. Antimicrob Agents Chemother.
tion of Rochalimaea henselae in cat scratch disease 1984; 25:690-3.
skin test antigens. J Infect Dis 1993;168:1034-6.

Dispatches
Emergence of Barmah Forest Virus

1
in Western Australia
To the editor: Barmah Forest (BF) virus is a obtained from mosquitoes trapped in several differ-
mosquito-borne alphavirus, found only in Australia, ent regions of Western Australia (Table 1). The first
which causes outbreaks of polyarthritis in humans. human cases of BF virus disease in Western Austra-
The disease is very similar to epidemic polyarthritis lia were reported in 1992, and 67 serologically con-
caused by infection with Ross River virus, another firmed cases have now been diagnosed. The
Australian alphavirus. BF virus was first isolated locations of towns where human cases have occurred
from mosquitoes in the State of Western Australia or where mosquitoes that yielded BF virus were
in 1989. After this, small clusters of human cases collected are shown in Figure 1.
were diagnosed in the arid northern and central Eight isolates of the virus were obtained from five
regions of Western Australia in 1992, and the first different mosquito species (Table 1) collected at Bil-
substantial outbreak of human disease due to infec- liluna, a small, remote aboriginal community in an
tion with BF virus (BF virus disease) occurred in the arid area in the southeastern Kimberley region in
southwestern region of the state during the spring April 1989 (10). The infected mosquitoes were col-
and summer (September-March) of 1993-94 (2). No lected 3 weeks after heavy local rains. Only moder-
evidence of BF virus activity had been found in these ate wet season rains were recorded in the remainder
regions before these events, which suggests that the of the Kimberley region, and no cases of BF virus
virus had only recently been introduced to Western disease were reported from any region in Western
Australia. This report describes the timing and dis- Australia that year. There have been no subsequent
tribution of BF virus disease in humans and the isolations of BF virus from mosquitoes collected at
isolation of the virus from mosquitoes in Western Billiluna, despite annual collections in the region.
Australia, which corroborate the view that BF virus No human cases have been reported from Billiluna.
is an emerging virus in this state. The first cases of BF virus disease in Western
The ecology of Australian arboviruses that cause Australian were reported almost 3 years later, either
human disease, including BF virus, has recently individually or in small clusters from towns in the
been reviewed (3). BF virus was first isolated from arid East Kimberley, Pilbara, Gascoyne, Murchison
Culex annulirostris mosquitoes collected at the Bar- and Southeast (Goldfields) regions between April
mah Forest in northern Victoria (southeastern Aus- and September (Autumn-Spring) 1992. Most activ-
tralia) in 1974 (4). It was first shown to infect ity was reported from the towns of Exmouth (six
humans in New South Wales (central-eastern and cases) and Carnarvon (four cases). All of these cases
southeastern Australia) in 1986 (5) and was reported occurred during or just after much larger outbreaks
as a cause of clinical disease in humans in 1988 (6). of disease caused by Ross River virus. This sug-
The most common clinical features include gested that BF and Ross River viruses may have
polyarthritis, arthralgia, myalgia, fever, rash, and similar mosquito vectors and require similar envi-
lethargy (7); in some cases, symptoms may persist ronmental conditions for successful transmission.
for more than 6 months (2). Although the symptoms The main environmental factor contributing to the
are similar to those caused by infection with Ross 1992 outbreaks of Ross River virus disease was
River virus, there is little cross-reaction between the extremely heavy rain in these normally arid regions
two viruses in serologic tests (8), and differentiating during autumn and winter (11).
between infections caused by either is generally not BF virus was isolated from five species of mos-
difficult. The first true outbreak of BF virus disease quito in the Fortescue region of the Pilbara and from
occurred concurrently with an outbreak of Ross three species in the West Gascoyne, just prior to, and
River virus infection at Nhulunbuy in the Northern during, these arid-region outbreaks. In coastal re-
Territory in early 1992 (9). gions of the Pilbara, the main vector of BF virus
The principal vectors of BF are believed to be appears to be Aedes vigilax, a salt marsh-breeding
mosquitoes, and although the vertebrate hosts of BF species. Large numbers of this species develop after
virus are not known, serologic surveys in eastern very high tides or heavy rains on salt marshes. It is
Australia have suggested that marsupials are in- also the main vector of Ross River virus in these
volved in the natural cycle. regions (12). Several other temporary freshwater
BF virus was first detected in Western Australia ground pool-breeding species in the subgenus
in 1989. Since then, 73 isolates of the virus have been Ochlerotatus, particularly Ae. eidsvoldensis and Ae.
EN Marks’ species #85, were found to be infected
1 with the virus in inland areas or coastal areas where
This report is adapted from and expands on a previous
bulletin. (1) such pools develop. These preliminary investiga-

Dispatches
Table 1. Mosquito species from which BF virus was isolated in Western Australia by region and
date, 1989–1993*
Region† Locality† Species Date Isolates
East Kimberley Billiluna Ae. bancroftianus 22 Apr 1989 1
East Kimberley Billiluna Ae. eidsvoldensis 22 Apr 1989 3
East Kimberley Billiluna Ae. pseudonormanensis 22 Apr 1989 1
East Kimberley Billiluna An. amictus 22 Apr 1989 2
East Kimberley Billiluna An. annulipes s.l. 22 Apr 1989 1
East Kimberley Halls Creek Cx. annulirostris 11 Feb 1993 1
West Kimberley Broome Ae. vigilax 10-16 Feb 1993 9
West Kimberley Fitzroy Crossing Cx. annulirostris 13 Feb 1993 1
West Kimberley Willare Ae. normanensis 16 Mar 1993 1
Pilbara (Fortescue) Onslow Ae. EN Marks’ sp. #85 13-14 Jun 1992 3
Pilbara (Fortescue) Onslow Cx. annulirostris 13 Jun 1992 1
Pilbara (Fortescue) Onslow An. amictus 13-14 Jun 1992 2
Pilbara (Fortescue) Exmouth Ae. vigilax 16 Jun-11 Jul 1992 7
Gascoyne (West) Minilya Ae. eidsvoldensis 7 Jul 1992 5
Gascoyne (West) Minilya Ae. eidsvoldensis (bloodfed) 7 Jul 1992 1
Gascoyne (West) Minilya Ae. EN Marks’ sp. #85 7 Jul 1992 1
Gascoyne (West) Carnarvon Ae. eidsvoldensis 12 Jul 1992 3
Gascoyne (West) Carnarvon Ae. EN Marks’ sp. #85 12 Jul 1992 1
Gascoyne (West) Carnarvon Cx. quinquefasciatus 12 Jul 1992 1
Gascoyne (West) Carnarvon Unidentifiable mosquitoes 12 Jul 1992 1
Central Coastal Karnup Cx. annulirostris 4 Jan 1993 1
Central Coastal Karnup Cq. species near linealis 4 Jan 1993 4
Central Coastal Perth Cx. annulirostris 6 Jan 1993 1
South Coastal Australind Ae. camptorhynchus 6 Jul 1993 1
South Coastal Mandurah (Peel) Ae. camptorhynchus 5 Aug-5 Oct 1993 10
South Coastal Busselton Ae. camptorhynchus 1 Sep-15 Nov 1993 9
South Coastal Busselton Cx. globocoxitus 1 Nov 1993 1
Total 73
* Numbers of mosquitoes trapped and processed and estimated minimum field infection rates for each region will be published in
detail elsewhere.
†
Refer to Figure 1 for location of regions and towns from which isolates of BF virus were obtained.
tions also suggested that both BF and Ross River locations, Table 1 for isolation details). A twelfth
virus can co-circulate. Both viruses were isolated isolate was obtained from Ae. normanensis collected
from different mosquitoes of the same species col- 5 weeks after the first rains at Willare in the West
lected in the same trap on several occasions. Kimberley. The timing of the collections was such
A further six cases of BF virus disease were re- that all three mosquito species could have transmit-
ported after record wet season rains in the Kimber- ted BF virus to the infected persons in the region.
ley region in early 1993. The cases occurred just after Vector competence studies are required to determine
mosquitoes in the Kimberley region had been col- if one or more species were likely to have been the
lected by personnel from this laboratory. These col- main vectors.
lections yielded 12 isolates of BF virus. Eleven of A single case was reported from the metropolitan
these were from Ae. vigilax and Cx. annulirostris area of Perth, the state’s capital, in August 1992.
mosquitoes trapped less than 2 weeks after the first This was the first evidence of BF virus activity in the
heavy wet season rains near the West Kimberley temperate and populous southwestern region of
towns of Broome and Fitzroy Crossing and the East Western Australia. However, the travel history of
Kimberley town of Halls Creek (see Figure 1 for the patient was not obtained. Then, in early January

Dispatches
was isolated from Ae. camptorhyn-

chus mosquitoes trapped in July
and August (Table 1).
A larger outbreak of BF virus
disease occurred in the southwest
region between September 1993
and March 1994. This has been de-
scribed in detail elsewhere (2).
Twenty-eight serologically con-
firmed cases were reported from the
southwest region during that pe-
riod. Of these, more than half (17
cases) were in or near the small
coastal towns of Mandurah (Peel)
and Busselton during spring (Sep-
tember-November) of 1993. BF vi-
rus was isolated on 20 occasions
from pools of Ae. camptorhynchus
mosquitoes collected in the Man-
durah and Busselton regions before
and during the outbreak (Table 1),
thereby implicating that species,
along with Cx. annulirostris and
Cq. species near linealis, as an im-
portant vector in the southwest. Ae.
camptorhynchus breeds in salt
marshes and brackish wetlands
during all but the hottest months of
the year in the southwest and is the
main vector of Ross River virus in
Figure 1. Meteorological regions and towns in Western Australia where the region (3, 11). The ratio of the
human cases of Barmah Forest virus disease were reported and Barmah
carriage rate of BF virus in Ae.
Forest virus was isolated from mosquitoes, 1989–1994. Meteorologic
regions: 1. North Kimberley; 2. East Kimberley; 3. West Kimberley; camptorhynchus during the out-
4. Pilbara (De Grey); 5. Pilbara (Fortescue); 6. West Gascoyne; 7. East break to the number of human cases
Gascoyne; 7A. Murchison; 8. North Coastal; 9. Central Coastal; 9A. South was very high compared with the
Coastal; 10. North Central; 10A. South Central; 11. Eucla; 12. Southeast rate observed for Ross River virus
(Goldfields); and 13. Northeast (Interior). in Ae. camptorhynchus in the same
regions during previous Ross River
1993, this laboratory isolated BF virus from Cx. virus outbreaks (M.D. Lindsay, C.A. Johansen, and
annulirostris and Coquillettidia species near J.S. Mackenzie, unpublished observations). This
linealis2 mosquitoes collected at Karnup, south of suggests that the ratio of subclinical (therefore un-
Perth (Table 1). A week later a single human case reported) to clinical cases may be much larger with
was reported from an address near the site at which BF virus than with Ross River virus or that fewer
the mosquitoes were trapped. BF virus was also humans were infected, possibly because Ae. camp-
isolated from Cx. annulirostris mosquitoes trapped torhynchus may not transmit BF virus to humans as
in the southern suburbs of Perth in early January efficiently as it does Ross River virus. Seven cases
1993, providing evidence that the virus may be were reported between November 1993 and March
transmitted to humans in the metropolitan area. 1994 in small towns in the inland southwest region
Two further cases were reported from the Perth in a later cycle of virus activity. Unfortunately, no
metropolitan area, one in February and one in May collections of mosquitoes were carried out in the
1993. Again, no travel histories were available for region during that time. Small clusters of cases or
these patients, but it appears that the virus re- individual cases were also reported from several
mained and was actively transmitted in the south- other regions of Western Australia during this time,
west during the autumn and winter of 1993, as it including three additional cases from Broome in the
West Kimberley region, presumably associated with
the 1993-94 wet season.
2
This species is similar to, but distinct from Coquillettidia BF virus disease was made a notifiable disease in
linealis, according to E.N. Marks, the leading Culicid taxono- Western Australia in June 1994 as a direct result of
mist in Australia.

Dispatches
the 1993-94 southwest outbreak (Health [Infectious particularly whether the strain of virus responsible
Diseases] Amendment Order 1994, Government Ga- for the Western Australia outbreaks was introduced
zette, Western Australia, 24 June 1994). The out- from Eastern Australia or was a local, hitherto un-
break was also the first report of a substantial detected, strain.
number of cases in the absence of Ross River virus
activity anywhere in Australia. Ross River virus is
endemic in the Mandurah (Peel) region but only one Acknowledgments
case of Ross River virus disease was reported from The authors thank Tony Wright, (Health Depart-
the area during spring-summer (September-Febru- ment of Western Australia) for assistance with col-
ary) 1993-94. This is the lowest recorded number of lection of mosquitoes, Margaret Wallace, Rosa
cases for that period in the region since record keep- Egerton-Warburton, Roger Clifton, and Dora Li for
ing began in 1984. Environmental conditions and assistance with laboratory processing of mosquito
vector mosquito populations in the southwest were and virus samples, and staff in the Virology Section,
unfavorable for Ross River virus transmission dur- State Health Laboratory Service, the Health De-
ing the BF outbreak. In particular, populations of Ae. partment of Western Australia, for carrying out the
camptorhynchus from October onwards were much serologic testing of human sera. This work was
smaller than in years when larger numbers of cases funded by a grant from the Health Department of
of Ross River virus disease were reported (M.D. Western Australia.
Lindsay, C.A. Johansen, J.S. Mackenzie, unpub-
Michael D.A. Lindsay, Cheryl A. Johansen,
lished observations). It is not known whether the BF Annette K. Broom, David W. Smith,* and
virus outbreak occurred because BF virus can circu- John S. Mackenzie
late under conditions that are not suitable for Ross Department of Microbiology, The University of Western
River virus activity or whether extremely low levels Australia, Queen Elizabeth II Medical Centre,
of immunity in “virgin” vertebrate host and human Nedlands, WA
populations in the southwest may have enhanced *State Health Laboratory Services, Queen Elizabeth II
transmission cycles. Medical Centre, Nedlands, WA
Surveillance and epidemiologic studies carried
out by this laboratory in the north of Western Aus- References
tralia since 1972 and in the southwest since 1987 1. Lindsay MD, Smith DW, Johansen C, Mackenzie JS.
have found no convincing evidence of BF virus activ- Barmah Forest virus disease in Western Australia.
ity in these regions prior to the events described in Western Australian Notifiable Diseases Bulletin 1994;
this report. No BF virus was isolated from the north 4: 1-4.
of Western Australia before 1989, despite large-scale 2. Lindsay MD, Johansen C, Smith DW, Wallace MJ,
processing of field-caught mosquitoes over a 17-year Mackenzie JS. An outbreak of Barmah Forest virus
period that yielded hundreds of isolates of other disease in the south-west of Western Australia. Med J
arboviruses. Similarly, no BF virus isolate was ob- Aust. In press.
tained from more than 400,000 mosquitoes collected 3. Mackenzie JS, Lindsay MD, Coelen RJ, Hall RA,
Broom AK, Smith DW. Arboviruses causing human
throughout the southwest between 1987 and 1992
disease in the Australian zoogeographic region. Arch
and processed for virus isolation. Furthermore, an Virol 1994;136:447-7.
ongoing serosurvey has found no evidence of infec- 4. Marshall ID, Woodroofe GM, Hirsch S. Viruses recov-
tion with BF virus in more than 1,000 individuals of ered from mosquitoes and wildlife serum collected in
18 vertebrate species collected in the southwest be- the Murray Valley of south-eastern Australia, Febru-
fore 1992 (C.A. Johansen, unpublished results). This ary 1974, during an outbreak of encephalitis. Aust J
suggests that the virus responsible for the recent Exp Biol Med Sci 1982;60:457-70.
outbreaks was recently introduced to Western Aus- 5. Vale TG, Carter IW, McPhie KA, James GS, Cloonan
tralia. The means of introduction, initially to the MJ. Human arbovirus infections along the south coast
northwest and more recently to the southwest of of New South Wales. Aust J Exp Biol Med Sci
1986;64:307-9.
Western Australia, is not known. In view of the
6. Boughton CR, Hawkes RA, Naim HM. Illness caused
activity at Nhulunbuy in the Northern Territory,
by a Barmah Forest-like virus in New South Wales.
before the first Western Australia cases in 1992, it Med J Aust 1988;148:146-7.
is possible that the virus may have been introduced 7. Phillips DA, Murray JR, Aaskov JG, Weimers MA.
from that region in a viremic human or in livestock. Clinical and subclinical Barmah Forest virus infection
However, little is known about the duration and in Queensland. Med J Aust 1990;152:463-6.
height of viremia in infected humans or other ani- 8. Hawkes RA, Boughton CR, Naim HM, Myrick BA,
mals, and it is not known whether person-to-person Ramsay LG. Barmah Forest virus infections in hu-
vector-mediated transmission of Barmah Forest vi- mans in New South Wales. Med J Aust 1988;146:569-
rus can occur. Our laboratory has begun a study to 73.
investigate the molecular epidemiology of BF virus,

Dispatches
9. Merianos A, Farland AM, Patel M, Currie B, Whelan 11. Lindsay MD, Johansen C, Wright AE, Condon R, D’Er-
P, Dentith H, Smith D. A concurrent outbreak of Bar- cole M, Smith D, Mackenzie JS. The epidemiology of
mah Forest and Ross River disease in Nhulunbuy, outbreaks of Ross River virus infection in Western
Northern Te rritory. Comm Dis Intell (Aust) Australia in 1991-1992. Arbovirus Research in Austra-
1992;16:110-1. lia 1992;6:72-6. CSIRO and Queensland Institute of
10. Broom AK, Mackenzie JS, Lindsay MD, Wright AE. Medical Research.
Epidemiology of Murray Valley encephalitis and Kun- 12. Lindsay MD, Broom AK, Wright AE, Johansen CA,
jin viruses in Western Australia, 1980-89. Arbovirus Mackenzie JS. Ross River virus isolations from mos-
Research in Australia 1989;5:14-8. CSIRO and quitoes in arid regions of Western Australia: implica-
Queensland Institute of Medical Research. tion of vertical transmission as a means of persistence
of the virus. Am J Trop Med Hyg 1993;49:686-96.
An Outbreak of Shigella sonnei Infection

Associated with Consumption of Iceberg Lettuce
To the Editor: Shigella sonnei outbreaks in CDSC were scrutinized to determine the age group
England and Wales are typically associated with and sex distributions during weeks 21 to 24.
primary schools and nurseries. The mode of trans- After the Epinet message, CsCDC and laboratory
mission is usually from person to person by the directors who reported clinical cases for which Sh.
fecal-oral route (1). In a June 1994 outbreak of Sh. sonnei was isolated were asked to administer trawl-
sonnei food poisoning among adults in several coun- ing questionnaires to apparently sporadic cases
tries in North West Europe, the vehicle of infection among adults with no recent history of overseas
appeared to be iceberg lettuce (2). travel. Personal details and history of illness and
In early June, the Communicable Disease Sur- exposure to particular foods were sought.
veillance Centre (CDSC), Public Health Laboratory Several small outbreaks and clusters were re-
Service, received a report of an increase in domestic ported during June. CsCDC was asked for results of
cases of Sh. sonnei infections in Sweden from the any analytical epidemiologic studies to CDSC. The
Salmnet network—a European international labo- results of the national laboratory reporting system
ratory-based reporting system for human salmo- are shown in Table 1. Although there were fewer
nella infections that provides a timely on-line reports in the first 20 weeks of 1994 than in a similar
database. In this instance the network was used for period in 1993, there were more reports in the weeks
shigellosis. Of 100 reported cases of Sh. sonnei infec- 21 to 24 and many more reports among adults. The
tion in Sweden, 52 occurred in two outbreaks in mid- proportion of total reports constituted by those from
May. Many cases seemed to be due to foodborne adults was 66% in weeks 21 to 24 of 1994 compared
infection, and iceberg lettuce and peeled frozen with 44% with the same period in 1993. The propor-
prawns were implicated as vehicles of infection. Sh. tion in women in the 2 periods was 42% in 1994
sonnei phage types 2 and 3 alpha were associated compared with 26% in 1993.
with the outbreaks, and phage types 2 and 65 had been Forty trawling questionnaires were distributed.
isolated from sporadic cases. Almost all case patients (38/40) had eaten various
A message was sent throughout England and salad items of which the common food was iceberg
Wales on Epinet (a system for rapid electronic data lettuce. The lettuce had been consumed in restau-
transfer to all Consultants in Communicable Dis- rants, pubs, and in the homes of the case-patients.
ease Control [CsCDC] in each District Health The lettuce was purchased from supermarkets,
Authority, Public Health Laboratories [PHLs] and greengrocers’ shops, and street markets. In one out-
other agencies involved in infectious disease control) break in Northampton, 21 (52%) of guests at a party
asking for information on possible foodborne Sh. became ill with diarrhea. Sh. sonnei was isolated
sonnei infection to be sent to CDSC and for isolates from fecal specimens. Illness was significantly asso-
to be referred to the Laboratory of Enteric Patho- ciated with consumption of iceberg lettuce (relative
gens (LEP) for phage typing. risk 3.68, confidence intervals 1.34 - 10.11, p =
0.0004).
The hypothesis that consumption of iceberg let-
Epidemiologic studies
tuce was associated with apparently sporadic Sh.
Laboratory reports of Sh. sonnei infection re- sonnei infection in adults was tested by a case-con-
ceived through the routine reporting system at

Dispatches
Table 1. Shigella sonnei in England and Wales—Laboratory reports to CDSC

Year Number of reports (%)
Week 1–20 Weeks 21–24 Weeks 1–24
Total Total Adults Women Total

1993 3190 480 (100) 211 (44) 127 (26) 3670
1994 1557 505 (100) 333 (66) 214 (42) 2062
trol study. A case was defined as a person aged 14 or porters supplying wholesale markets in England.
more years who became ill after May 1, 1994, and The wholesalers were supplied by packers in Spain.
had microbiologic evidence (fecal isolation) of Sh. This was consistent with the findings of the investi-
sonnei infection, no recent history of overseas travel, gators in the Norwegian outbreak. Iceberg lettuce
and no identifiable contact with other case-patients investigated by the Public Health Laboratory serv-
in the 3 days before onset. Controls were nominated ice during the second week of June 1994 did not grow
by case-patients and matched by sex, age (within a Sh. sonnei. However, the iceberg lettuce season in
10-year age band), and area of residence (within a Spain, which began in October, ended early in June,
10-mile radius of the case). For each case three and the source of lettuce available for testing could
matched controls were sought. A questionnaire was not be traced.
administered by telephone by three interviewers
from CDSC. Clinical and demographic details and
details of exposure to food items, including iceberg Laboratory studies
lettuce, mentioned in trawling questionnaires were All Sh. sonnei isolates referred to LEP after the
sought. Epinet message were phage typed by using the
Twenty-eight case-patients and 49 matched con- scheme described by Hammerstrom (3) and Kallings
trols were interviewed and, after excluding those and Sjoberg (4), according to a protocol supplied by
who had recently traveled abroad and controls who Dr. R. Wollin, Swedish Institute of Infectious Dis-
had been ill, results from 27 cases and 44 controls ease Control, Sweden. Isolates were also tested for
were analyzed. The median age of case-patients was resistance to a range of antibiotics by an agar dilu-
47 years, and the range was 19 to 79 years. Eight tion technique (5).
cases were among men and 19 among women. All A total of 495 isolates were referred to LEP be-
case-patients had diarrhea (i.e., three or more loose tween June 14 and July 31, from 51 laboratories in
stools in a 24-hour period), although only four of the England and Wales. Most isolates were from spo-
27 reported blood in the stools, 25 of the 27 had radic infections, but in a number of local outbreaks,
abdominal pain, and 11 reported vomiting. The me- there was a strong epidemiologic association be-
dian duration of symptoms was 9 days, and the tween illness and consumption of iceberg lettuce.
range was 4 to 25 days. Taking into account the Two phage types predominated among the 19 types
matching inherent in the study design, a matched identified during this period, PT 2 (42.6% of isolates)
analysis was performed. In any analysis 27 matched and a variant of PT65 provisionally designated PT
sets were possible. For 13 sets there was one control L (15.9%). In contrast, although a small number of
per case, for 11 sets there were 2 per case, and for 3 isolates of PT 65 and PT L were identified among
sets there were 3 controls per case. Single variable strains of Sh. sonnei isolated in England and Wales
analysis of the different foods consumed revealed the in 1991 and 1992, no isolates of PT 2 were seen
possible risk factors (p < 0.2) (Table 2). A multivari- before May 1994. Towards the end of the outbreak
able model was fitted with all those variables in- PT 3 and PT 6 were becoming reestablished in Eng-
cluded. This procedure was repeated, removing land and Wales as the predominant types, as they had
nonsignificant items at each stage. In the third been in previous years.
model, the only remaining significant item was ice- An exception to the recent pattern was an out-
berg lettuce (p = 0.0172). The estimated odds ratio break in North Wales, involving several children and
for iceberg lettuce was 13.8 (95% confidence interval adults, in which infection was associated either with
1.26 to 150.5). eating ice cream at a particular establishment or
In sporadic cases associated with consumption of having contact with children who had done so. All 73
lettuce from particular restaurants or public houses, of the isolates of PT 62 were associated with this
it was possible to compare the date of onset with the outbreak.
date of delivery of iceberg lettuce by the wholesalers. A total of 357 Sh. sonnei isolated during this
The distribution chain was traced back through im- period (72.1%) were fully sensitive to all drugs

Dispatches
Table 2. Sh. sonnei case control study—Single variable matched analysis

Case (n=27) Control (n=44)
Did not Did not No. of
Food Ate eat % ate Ate eat % ate sets p-value
Prawns 4 22 15 7 37 15 (26s) 0.9388
Shrimps 0 27 0 1 43 2 (27s) 0.3070
Steak 7 19 27 6 37 14 (26s) 0.0139
Burgers 5 22 19 12 32 27 (27s) 0.2045
Salad 25 2 93 34 10 77 (27s) 0.0081
Cold meats 15 10 60 29 13 69 (23s) 0.7708
Tomatoes 22 4 85 33 11 75 (26s) 0.1039
Spring onions 5 21 19 9 35 21 (26s) 0.7556
Celery 5 22 19 5 39 11 (27s) 0.2956
Cucumber 16 11 59 31 13 71 (27s) 0.3798
Other salads 16 10 62 26 18 59 (26s) 1.0000
Lettuce 25 2 93 31 13 71 (27s) 0.0007
Cos lettuce 2 20 9 3 25 11 (17s) 1.0000
Webb’s lettuce 5 18 22 4 34 11 (21s) 0.2632
Lamb’s lettuce 2 17 11 6 36 14 (19s) 0.1002
Raddicio lettuce 4 20 17 7 34 17 (23s) 0.8720
Iceberg lettuce 17 8 68 19 23 45 (25s) 0.0023
Frisee lettuce 5 18 22 4 37 10 (22s) 0.2582
Home 15 8 65 27 16 63 (23s) 0.8494
Restaurant 6 16 27 5 33 13 (21s) 0.1546
Pub 4 18 18 2 34 6 (20s) 0.0795
Other outlet 10 14 42 7 31 18 (22s) 0.0012
tested. Phage types 2 (87% fully sensitive) and PT L infections in Norway and Sweden (6). A plausible
(99%) were predominantly sensitive, as were all explanation would be that fecally contaminated
isolates of PT 62; usually one would expect more water was used to irrigate the lettuce or to cool it
than 70% of Sh. sonnei isolates to be resistant to one after packing. If iceberg lettuce is not washed thor-
or more drugs. The use of the same phage-typing oughly before consumption, contamination could be
scheme across several European countries has facili- retained in the leaves.
tated cross-referencing between the British, Ger- This study demonstrates both the importance of
man, and Swedish outbreaks. Phage types 2 and 65 coordinating laboratory results and epidemiologic
(or the closely related variant PT L) were identified investigations and the value of rapid communica-
in several countries. tions and common typing techniques in various
From the epidemiologic studies, it was concluded European countries.
that the strong statistical evidence (p = 0.0172) that
consumption of iceberg lettuce was associated with J.A. Frost, M.B. McEvoy,* C.A. Bentley,
Y. Andersson,** and B. Rowe
the risk of becoming ill together with reports from Laboratory of Enteric Pathogens,
other European countries, including Scotland, Swe- Central Public Health Laboratory, London, England
den, and Norway, and the temporal association of the *Communicable Disease Surveillance Centre,
outbreak with the iceberg lettuce season in Spain London, England
implicated iceberg lettuce as the vehicle of infection. **Swedish Institute for Infectious Diseases Control,
This was corroborated by the laboratory studies, Stockholm, Sweden
which showed a change in predominant phage types
during the period of the outbreak. The predomi-
nance of the same phage types in lettuce-associated References
Sh. sonnei infections in a number of countries added 1. Newman CPS. Surveillance and control of Shigella
sonnei infection. Communicable Disease Report 1993;
further weight to this conclusion. 3: R63-8.
In England, there were several anecdotal ac- 2. Anonymous. A foodborne outbreak of Shigella sonnei
counts of dual infection with salmonellae and vi- infection in Europe. Communicable Disease Report
ruses as well as Sh. sonnei. This was also true of 1994; 4 No. 25.

Dispatches
3. Hammerstrom E. Phage typing of Shigella sonnei. Acta 5. Frost JA. Testing for resistance to antimicrobial drugs.
Med Scand 1949. 133 (Suppl 223). In: Chart H, ed. Methods in practical laboratory bacte-
4. Kallings LO, Lindberg AA, Sjoberg L. Phage typing of riology. Boca Raton Fla.: CRC Press; 1994.
Shigella sonnei. Arch Immun Ther Exp 1968; 16: 280-7. 6. Kapperud G, Rorvik LM, Hasseltvedt V, et al. Out-
break of Shigella sonnei infection traced to imported
iceberg lettuce. J Clin Microbiol, in press.
?Lyme Disease in Australia— Still To Be Proven!

To the Editor: The first case of a syndrome acrodermatitis chronica atrophicans, seen occasion-
consistent with Lyme disease was reported from the ally in Europe and rarely in the United States, has
Hunter Valley region of New South Wales (NSW) in not been reported from Australia.
southeastern Australia in 1982, but there was no A few cases of erythema migrans, the charac-
confirming serology. More clinical cases, again with- teristic dermatologic manifestation of acute Lyme
out serologic confirmation, were reported in 1986, disease, have been reported from southeastern Aus-
two from the south coast and one from the central tralia, but clinical diagnosis can be confounded by
coast of NSW. The Queensland State Health Labo- hypersensitivity reactions to tick bite; a spectacular
ratories reported that 186 (14.9%) of 1,247 sera erythematous reaction is often associated with the
taken from patients between 1986-1989 showed bite of Ixodes holocyclus, the most common tick
antibody response to Borrelia burgdorferi of ≥ 64 by biting humans in NSW. Only eight specimens sub-
indirect fluorescence antibody test (IFAT), but none mitted to our laboratory included skin biopsies done
of these results were confirmed by immunoblotting. to isolate spirochetes. B. burgdorferi s.l. was iso-
In 1988, a multidisciplinary investigation of pu- lated from one patient returning from Europe, but
tative Lyme disease began, encompassing clinical, no spirochetes were isolated from local patients.
serologic, vector, and reservoir host studies, and In our serologic diagnostic service, an enzyme-
results from these studies have been published (1). linked immunosorbent assay (ELISA) for IgG and
What follows herein is derived from the accumu- an IFAT for IgG and IgM have been used with
lated published and unpublished data of the re- antigens derived from North American B. burgdor-
search team, the members of which are credited in feri strain B31 (2). From 1988 to April 1994, 78
the acknowledgments. (1.8%) of 4,372 local patients were positive for IgG
Over the past 6 years, principally because of local by both methods. All 78 patients were tested by IgG
publicity, there has been an increase in serologic Western blot for confirmation by using the virulent
testing for Lyme disease in Australia, particularly North American B. burgdorferi strain 297 and a
in southeastern Australia. Testing has often been German strain designated B7: with B. burgdorferi
initiated by patients with undiagnosed health prob- strain 297, 46 patient samples showed as many as
lems. Thus, most Lyme disease patients seen by four indicative bands; with the European strain B7,
infectious disease specialists are self selected and 22 patient samples showed as many as three indica-
are referred for assessment on the basis of tick tive bands; bands used were 18, 21, 28, 30, 31, 34,
exposure and reported positive serologic test results 39, 41, 45, 58, 66, 83, and 93 kDa, modified from
for Lyme disease. Dressler et al (3). Twenty-four other patients with
Patients with positive serologic test results fre- various bacterial, viral, or autoimmune syndromes
quently have long-standing symptoms for which no not relating to Lyme disease were tested as controls:
other diagnosis has been established. The most com- with strain 297, 11 control samples showed as many
mon symptoms are musculoskeletal, including my- as two indicative bands, and with strain B7, 10
algias and arthralgias without objective evidence of control samples showed as many as two indicative
joint swelling, and syndromes involving fatigue and bands.
loss of energy resembling chronic fatigue syndrome. A high degree of cross-reactivity was demon-
Some patients fulfill diagnostic criteria for fibromy- strated with the controls, particularly with respect
algia. The next most common symptoms are neuro- to the 31, 41, 58, and 66 kDa bands for both the
logical, and include frequent headaches, inability to European and the American antigen. As none of the
concentrate, and memory loss. The most common 78 patients, including putative late-stage patients
dermatologic manifestation of chronic Lyme disease, positive by ELISA and IFAT, showed more than four

Dispatches
specific bands to either antigen, they would be contaminants. Molecular characterization indicated
sidered negative by the criteria of Dressler et al (3). that the SLOs were not related to B. burgdorferi.
Fewer than 1% of all referred patients conformed A small number of native vertebrate animals (13
with the national surveillance case definition used native rats Rattus fuscipes, 3 bandicoots Perameles
in the United States by the Centers for Disease nasuta, and 1 marsupial mouse Antechinus stuartii)
Control and Prevention. Problems of specificity and trapped on the south coast of NSW were sampled by
sensitivity associated with serologic testing for ear-punch biopsy (6) for culture and PCR investiga-
Lyme disease are well recognized, particularly in tion, but no evidence of borreliae was found. The
Australia where no local spirochete has been iso- animal sample was clearly inadequate, and the PCR
lated for use as a reference antigen. primers used for the tick and animal studies may
Seroprevalence rates for B. burgdorferi infection have been inappropriate and unable to identify na-
in humans have been compared between 200 high tive Australian spirochetes; however the extensive
(rural residents) and 200 low (urban residents) tick investigations of tick gut contents by culturing and
exposure groups in coastal NSW, by using the IgG dark field microscopy were also negative for borre-
ELISA. No significant difference was found between liae.
the two groups, and the overall seropositivity rate There are some major differences between Austra-
was 2.2% (9/400). A parallel survey of dogs in NSW lia and the Lyme-disease-endemic areas of the North-
has shown a similar result with an overall seroposi- ern Hemisphere with respect to the natural history of
tivity rate of 2.5% (6/239). These results contrast borreliosis. No ticks of the I. persulcatus complex, the
with those reported from known endemic-disease principal vectors to humans in the northern hemi-
areas outside Australia that have rural populations sphere, occur in Australia. In eastern Australia, the
with considerably higher seropositive rates. The low logical candidate vector would be I. holocyclus,
rate found by our surveys is similar to that found by which has a wide host range and is the most common
other studies undertaken in areas where Lyme dis- tick biting humans. I. holocyclus cannot transmit a
ease is not endemic, and humans have 1%-3% posi- North American strain of B. burgdorferi (7) but the
tive serologic results cause by cross-reacting association with any possible Australian spirochetes
antibodies (4). remains unresolved. Likewise, none of the mammal
From January 1990 to December 1992, ticks were species identified as reservoir hosts in the Northern
collected in areas associated with putative Lyme Hemisphere are present in Australia. There are
disease cases and were examined for spirochetes to reports of spirochetes in Australian native animals,
detect a possible causative agent in potential vec- and a local mammal could be a reservoir host for an
tors. Ticks were collected along the east coast of indigenous spirochete that occasionally infects hu-
Australia, from southern Queensland through NSW mans through a tick vector and produces a clinical
into northern Victoria, by flagging in natural habi- syndrome similar to Lyme disease; however, no spi-
tats, and from domestic and other native animals. rochete was detected in the ticks or animals studied.
Detection of spirochetes was attempted by dark-field The diagnosis of Lyme disease outside known
microscopy and culturing of gut contents and by disease-endemic areas should not be based solely on
direct testing of ticks using polymerase chain reac- serology because unrelated syndromes, such as
tion (PCR) to detect the Borrelia-specific flagellin autoimmune diseases and cross reactions with other
gene (5). bacteria, can produce false-positive results. Like-
In total, more than 12,000 (>1,000 by PCR) ticks wise, a definitive diagnosis on clinical grounds alone
were processed, including 7,922 I. holocyclus (1). No in a nonendemic-disease area is difficult to justify
spirochetes were detected by dark-field microscopy without adequate scientific support based on isola-
or by PCR. Spirochete-like objects (SLOs), were ob- tion of the causative agent from the patient or from
served in 94 cultures from bloodfed ticks and only in another patient or known vector from the region. In
cultures with bacterial contaminants, presumably Australia, disagreement as to what constitutes a
from the bloodmeal. Some SLOs yielded positive positive serologic result has additionally contrib-
fluorescence results when tested with Borrelia-spe- uted to overdiagnosis of Lyme disease. Until an
cific polyclonal antibodies, but tests with monoclonal organism is isolated from a local patient and is
antibodies (anti-flagellin H9724, anti-OspA H5332, characterized, the presence of Lyme disease in Aus-
anti-OspB H6831) were negative. Electron mi- tralia will remain controversial.
crographs showed that the SLOs were not typical of
Borrelia, were composed of fibers, and probably were
not spirochetes. The electron micrographs were Acknowledgments
similar to micrographs of SLOs recovered from con- My colleagues Rosemary Munro (clinical microbi-
taminated cultures from ticks in the United States ology), Department of Microbiology, Liverpool Hos-
and Europe and thought to be composed of aggrega- pital, and Stephen Doggett (entomology), David
tions of bacterial flagella, probably from the con- Dickeson (serology), Danielle Avery (molecular biol-

Dispatches
ogy), Cheryl Hunt (molecular biology), Joanne Mer- References

cer (microbiology), and Nicole Trivett (electron mi- 1. Russell RC, Doggett SL, Munro R, Ellis J, Avery D,
croscopy), CIDM at Westmead Hospital, and John Hunt C, Dickeson D. Lyme disease: a search for the
Ellis (molecular biology), Department of Microbiol- causative agent in ticks in southeastern Australia.
ogy, University of Technology Sydney, contributed to Epidemiol Infect 1994;112:375-84.
the studies outlined in this paper. Richard 2. Russell H, Sampson JS, Schmid GP, Wilkinson HW,
Lawrence, Clinical Superintendent of Medicine, Plikaytis B. Enzyme-linked immunosorbent assay
and direct immunofluorescence assay for Lyme dis-
Westmead Hospital, provided valuable discussions
ease. J Infect Dis 1984;149:465-70.
on clinical aspects and case presentation. Our inves-
3. Dressler F, Whalen JA, Reinhardt BN, Steere AC.
tigations were supported by the National Health and Western blotting in the serodiagnosis of Lyme disease.
Medical Research Council and the Ramaciotti Foun- J Infect Dis 1993;167:392-400.
dations. 4. Barbour AG, Fish D. The biological and social phe-
nomenon of Lyme disease. Science 1993;260:1610-6.
Richard C. Russell
5. Persing DH, Telford III SR, Rys PN, Dodge DE, White
Department of Medical Entomology, Centre for
TJ, Malawista SE, Spielman A. Detection of Borrelia
Infectious Diseases and Microbiology,
burgdorferi DNA in museum specimens of Ixodes dam-
University of Sydney and Westmead Hospital,
mini ticks. Science 1990;249:1420-3.
Westmead, Australia
6. Sinsky RJ, Piesman J. Ear punch biopsy method for
detection and isolation of Borrelia burgdorferi from
rodents. J Clin Microbiol 1989;27:1723-7.
7. Piesman J, Stone BF. Vector competence of the Austra-
lian paralysis tick, Ixodes holocyclus, for the Lyme
disease spirochaete Borrelia burgdorferi. Int J Parasi-
tol 1991;21:109-11.
A Novel Morbillivirus Pneumonia of Horses and

its Transmission to Humans
To the editor: On September 22 and 23, 1994, tomatic. Nine Hendra horses have remained unaf-
veterinary authorities in Queensland and at the fected. The sick horses were anorexic, depressed,
CSIRO Australian Animal Health Laboratory were usually febrile (temperature up to 41°C), showed
advised of an outbreak of acute respiratory disease elevated respiratory rates, and became ataxic. Head
in horses at a stable in the Brisbane suburb of pressing was occasionally seen, and commonly, a
Hendra. The trainer of the horses had been hospi- frothy nasal discharge occurred before death.
talized for a respiratory disease and was in critical On September 14, a stablehand at the Hendra
condition. At that time, the cause of the horses’ stable developed an influenza-like illness charac-
illness was unclear and any link between equine and terized by fever and myalgia. The next day, the horse
human disease was thought improbable. Poisoning, trainer also became ill with similar symptoms. Both
bacterial, viral, and exotic disease causes were in- had close contact with the dying mare, particularly
vestigated. The history of the horses on this property the trainer who was exposed to nasal discharge
was considered important (Figure 1). Two weeks while trying to feed her; he had abrasions on his
before the trainer’s illness, on September 7, two hands and arms. The stablehand, a 40-year-old man,
horses had been moved to the Hendra stable from a remained ill for 6 weeks and gradually recovered.
spelling paddock in Cannon Hill (6 km). One of Besides myalgia, he also had headaches, lethargy,
these, a pregnant mare, was sick and died within 2 and vertigo. The trainer, a 49-year-old man, was a
days. The other horse was subsequently moved on heavy smoker and showed signs consistent with
and never became sick. By September 26, 13 horses Legionella infection. He ultimately required venti-
had died: the mare; 10 other horses in the Hendra lation for respiratory distress and died after 6 days
stable; one horse, which had very close contact with (Selvey L, et al. A novel morbillivirus infection caus-
horses in the Hendra stable, on a neighboring prop- ing severe respiratory illness in humans and horses,
erty; and one which had been transported from the submitted).
stable to another site (150 km). Four Hendra horses At the beginning of the diagnostic investigation
and three others (one in an adjacent stable, one in horses, African horse sickness, equine influenza,
moved to Kenilworth, and one to Samford) were and hyperacute equine herpes virus were excluded
later considered to have been exposed and recovered as possible causes by antigen trapping enzyme-
from the illness. Some of these horses were asymp- linked immunosorbent assay (ELISA), polymerase

Dispatches
September 1994
7 9 13 14 15 16 17 19-26
Horses
Cannon Hill 2 horses
(Paddock) moved
Hendra Mare died 2 horses 10 horses dead
(Stables) moved 4 recovered
Hendra 1 horse 1 horse dead
(Neighboring moved 1 recovered
property)
Kenilworth 1 horse dead
(150 km distant) 1 recovered
Samford 1 recovered
(Paddock)
New South Wales
Humans
Stablehand Becomes Slow recovery
ill
Trainer Becomes Hospitalized Died
ill
Figure 1. A chronology of the development of cases of acute equine respiratory disease and associated illness in
humans.
chain reaction (PCR), or electronmicroscopy. Tests PCR primers were synthesized from consensus
for Pasteurella, Bacillus anthracis, Yersinia, Le- Paramyxoviridae matrix protein sequences and
gionella, Pseudomonas, and Streptobacillus monili- tested against the horse virus. Those specific for
formis were negative, and poisons consistent with paramyxoviruses and pneumoviruses did not bind,
the clinical and gross pathology, such as paraquat, but one pair of morbillivirus primers gave a 400 bp
were excluded by specific testing. product. Determination of the sequence of this prod-
However, within 3 days, a syncytial forming virus uct enabled the synthesis of horse virus-specific
was detected in vero-cell cultures inoculated with primers. Phylogenetic analyses of the matrix protein
diseased horse tissues and shortly thereafter was sequence indicates that the virus is unique and
seen to grow in a wide range of cells. These included distantly related to other known members of the
MDBK, BHK, and RK13 cells. Subsequently, a group. A comparison of translated M protein se-
syncytial forming virus also was isolated in LLK- quence shows that it has a 50% homology with the
MK2 cells that had been inoculated with tissue from morbillivirus group (80% if conservative amino acid
the deceased trainer’s kidney. The isolation of these substitutions are used). This distant relatedness is
viruses and their preliminary characterization by emphasized by our observations that neutralizing
electron microscopy, immunoelectromicroscopy, se- antisera to measles virus, canine distemper, and
rology, and genetic analyses are described elsewhere rinperest virus failed to neutralize the virus.
(Murray PK, et al. A new morbillivirus which caused The viruses isolated from the horses and the
fatal disease in horses and man, submitted). trainer are ultrastructurally identical. Serum from
In summary, ultrastructural analysis showed the horses and the two human cases specifically
that the virus is a member of the Paramyxoviridae cross-neutralize the virus, and the horse virus-spe-
family. It is enveloped, pleomorphic (varies in size cific PCR primers provide a positive reaction with
from 38 nm to more than 600 nm), and is covered the human virus isolate. Therefore, the horses and
with 10 nm and 18 nm surface projections. It con- the trainer were infected with the same virus.
tains herringbone nucleocapsids that are 18 nm At the beginning of the diagnostic investigation,
wide with a 5 nm periodicity. The presence of ‘dou- tissues from the lungs and spleens of diseased horses
ble-fringed’ surface projections on this virus is con- were injected into two recipient horses. After 6 and
sidered unique. Immunoelectronmicroscopy showed 10 days, the recipient horses became ill with high
that both the horse and the human virus react with fever and severe respiratory signs, demonstrating
convalescent-phase horse sera and with sera from that the disease was transmissible. Two days later
the two human cases. the horses were destroyed. The equine morbillivirus

Dispatches
was isolated from tissues from both of these horses. sick horses, mostly stable workers, veterinary pa-
To document that the isolated horse virus was patho- thologists, animal health field staff, or people who
genic, experimental transmission tests were also lived in the vicinity of the affected stables, was
conducted. Two additional horses received a total of negative (Selvey L, et al, submitted).
2 x 107 TCID50 of tissue culture virus by intravenous Serologic testing of all horses on quarantined
inoculation and by intranasal aerosol. Both horses properties and within 1 km of the Hendra stable, and
became seriously ill, and after a short, severe clinical a sample of horses from the rest of Queensland was
episode, were destroyed 4 and 5 days after exposure. undertaken (Table 1). A total of 1,964 horses were
At necropsy, they showed gross and histopathologic tested from more than 630 premises. The negative
lesions that were primarily respiratory and consis- results from this testing also indicate that the infec-
tent with the natural disease. Virus was reisolated tion has not spread. In the entire horse survey, only
from their lungs, liver, spleen, kidney, lymph nodes, seven horses, all from the Hendra property and the
and blood. adjoining stables, were positive. Four of these ani-
The pathology of this infection is interesting. In mals had been clinically affected, but three were
horses, the dominant gross pathology is lesions in asymptomatic. Because of the potential risk and the
the lungs. These are congested and edematous with difficulty in establishing freedom from infection,
prominent lymphatic dilation in the ventral mar- these seven recovered horses were later destroyed.
gins. In natural cases, the airways were usually Although persistent virus excretion or carrier
filled with thick, fine, stable foam which was occa- states are not known to occur in other morbillivirus
sionally blood-tinged; this was not seen in the experi- infections, this equine virus is unique and it cannot
mental cases. Histologically, in horses, there is be presumed to behave similarly. Australian veteri-
interstitial pneumonia, proteinaceous edema with nary authorities are now satisfied that the incident
pneumocyte, and capillary degeneration. Virus can is over.
be located in endothelial cells by immunofluores- We have described a newly recorded disease of
cence and syncytial cells also could be seen in blood horses with an obvious zoonotic potential; moreover,
vessel walls, confirming the vascular of trophism of the causative agent was previously unrecorded and
this virus (Murray PK, et al, submitted). The is significantly different from other members of its
trainer’s post-mortem findings showed similarities genus, morbillivirus. Infection seems to have spread
to those of the horses (Selvey L, et al, submitted). from the mare that first showed the now charac-
No further clinical cases of disease have been seen teristic clinical signs, to other horses in the same
in horses or humans since this outbreak. Serologic stables, to a horse in close contact from an adjacent
surveillance of people who had close contact with the stable, and also to two human attendants. Clearly,
this outbreak was not highly contagious and it rap-
Table 1. The premises and horses surveyed idly resolved. However, the virus is highly patho-
by serologic testing for equine genic with 65% of naturally infected horses and all
morbillivirus, after the disease outbreak four experimental horses dying.
Further investigations of the virus and the dis-
Premises Horses
ease are now warranted since it could reemerge in
Quarantine Premises* 13 107 Australia or elsewhere. Investigations of its origin,
1 (within 100 m of 7 54 its replication, its pathogenesis, and its possible
Hendra stables) occurrence elsewhere in connection with equine res-
2 (100 m to 200 m of 21 122 piratory disease are merited.
Hendra stables)
Keith Murray, Russell Rogers,* Linda Selvey,**
3 (200 m to 1 km of 93 730 Paul Selleck, Alex Hyatt, Allan Gould, Laurie
Hendra stables) Gleeson, Peter Hooper, and Harvey Westbury
4/5 (remainder of >500 963 CSIRO Australian Animal Health Laboratory, Ryrie
Queensland) Street, East Geelong, Victoria 3220, Australia
Total >630 1,964 *Animal Health Bureau, Queensland Department of
Primary Industries, 80 Ann Street, Brisbane, Australia
* Quarantine premises included those with clinical cases, hold-
**Communicable Diseases, Queensland Health, 160
ing properties associated with the Hendra stables, and other
premises where horses under investigation were kept.
Mary Street, Brisbane 4000, Queensland, Australia

Commentary
Electronic Communication and in each district health authority, to the 53 public

health laboratories, and to other agencies in England
the Future of International and Wales involved in infectious disease control.
Further information from Norway, Scotland, and
Public Health Surveillance Sweden reinforced and stimulated the ongoing in-
Recent developments in electronic communica- vestigation in England and Wales. Since its incep-
tion have enhanced national public health surveil- tion, Salmnet has also contributed information with
lance systems and facilitated progress in potential international implications concerning sev-
establishing surveillance that crosses national eral Salmonella serotypes ( 3,4).
boundaries. The international foodborne outbreak of The signing of the Treaty of Maastricht (1992) was
Shigella sonnei described by Frost and colleagues an important milestone for international coopera-
was first reported through Salmnet (1), a laboratory- tion in public health surveillance. The treaty estab-
based surveillance system designed to include an lished a basis for European Community action in the
on-line network database. Salmnet was established field of public health and enjoined cooperation be-
in 1994 to improve the prevention and control of tween member states, third countries, and interna-
human salmonellosis and other foodborne infections tional public health organizations to protect human
in the countries of the European Union and the health. Sufficient evidence has accrued on the added
European Cooperation in Science and Technology. value of international surveillance of infectious dis-
Epidemiologists who have national surveillance eases, and it is generally accepted that the potential
responsibilities and heads of reference laboratories for major public health hazards is amplified as a
in 13 countries currently collaborate in the Salmnet consequence of the increasing volume of interna-
project under the joint leadership of the Directors of tional travel and the global extension of food distri-
the Laboratory of Enteric Pathogens and the Com- bution networks (5). Surveillance systems such as
municable Disease Surveillance Centre (CDSC) at Salmnet and the European Surveillance of Travel-
the Public Health Laboratory Service campus at Associated Legionnaires’ Disease (6) will pave the
Colindale, London. During the course of two work- way for international surveillance by providing a
shops, the collaborators agreed to a) develop and communications network that will facilitate the
apply standardized phage typing for the most com- rapid collection and analysis of data using standard
mon salmonella serotypes within Europe, b) intro- case definitions, transmission of information for the
duce an international quality assurance scheme for prevention of communicable diseases, and the pro-
laboratory performance of Salmonella phage typing, motion of effective public health practice.
c) establish a core set of data on each laboratory-con- Parallel electronic surveillance systems in the
firmed and typed human salmonella isolate for rapid United States offer equal opportunity for interna-
transfer into a shared nonaggregated dataset, and tional collaboration. For example, since 1985, data
d) develop statistical analysis programs to facilitate on notifiable diseases have been transmitted elec-
the early recognition of international outbreaks. The tronically each week to the Centers for Disease Con-
collaborators also agreed to rapidly report any clus- trol and Prevention (CDC) from state health
ters detected and to exchange information concern- departments by the National Electronic Telecommu-
ing other infections, including those caused by nications System for Surveillance (NETSS) (7).
Shigella, Listeria, and vero-cytotoxin-producing NETSS was developed by CDC and the Council of
strains of Escherichia coli. State and Territorial Epidemiologists for electroni-
Currently six countries share data through the cally collecting, transmitting, analyzing, and pub-
Internet, while the other seven countries rely on lishing weekly reports of notifiable diseases and
faxing material or sending diskettes through the injuries from 50 states, New York City, the District
mail. All collaborating countries plan to join the of Columbia, Puerto Rico, the Virgin Islands, Guam,
on-line network by the end of 1995. The opportuni- American Samoa, and the Commonwealth of the
ties offered by electronic communication have en- Northern Mariana Islands (8). The operation of
couraged a remarkable degree of international NETSS is based on agreements on reporting condi-
cooperation in surveillance, as is evident in the far tions, standard case definitions, and protocols for
reaching objectives agreed upon for Salmnet. formatting and transmitting data, rather than pre-
The Salmnet collaboration was already in place scribed software or systems.
when the S. sonnei outbreak in Sweden was reported A second CDC electronic reporting system, the
to Colindale. CDSC responded by sending an elec- Public Health Laboratory Information System
tronic message throughout England and Wales by (PHLIS), is used by public health department labo-
Epinet, an electronic system for the rapid transfer of ratories in all states, New York City, the District of
vital public health information developed by the Columbia, and Guam to report laboratory isolate-
CDSC Welsh Unit in Cardiff (2). The message was based surveillance data to CDC. The PC-based sys-
sent to consultants in communicable disease control tem was developed jointly by the National Center for

Commentary
Infectious Diseases, CDC, and the Association of References

State and Territorial Public Health Laboratory Di- 1. Fisher IST, Rowe B, Bartlett C, Gill O Noël."Salm-Net"
rectors to maintain a fast and direct link between laboratory-based surveillance of human salmonella in-
public health laboratories in the United States and fections in Europe. PHLS Microbiology Digest 1994;
CDC. PHLIS is used to gather, analyze and transmit 11:181-2.
data (e.g., laboratory testing results, epidemiologic 2. Palmer S, Henry R. Epinet in Wales: PHLS Cadwyn
information, findings from special studies and sur- Cymru development of a public health information
system. PHLS Microbiology Digest 1992;9:107-9.
veys) among multiple sources of public health labo-
3. Communicable Disease Surveillance Centre. Salmo-
ratory information (e.g., hospitals, laboratories, or
nella mikawasima update. Communicable Disease Re-
public health departments), and it provides an auto- port 1993;3:215.
mated program in its longitudinal databases to de- 4. Communicable Disease Surveillance Centre. Salmo-
tect outbreaks (9). nella javiana in Europe. Communicable Disease Report
European and U.S. surveillance databases and 1994;4:61.
information systems should be linked to share public 5. Bartlett C, Gill N. International surveillance of dis-
health information of international concern. To that ease—communicable disease control after Maastricht:
end, CDSC and CDC are developing a cooperative Germs and subsidiarity. Lancet 1993; 341:997-8.
communications information system that will use 6. Communicable Disease Surveillance Centre. European
the Internet to mirror vital public health documents surveillance of Legionnaires’ disease associated with
travel. Communicable Disease Report 1994;4:25.
(e.g., CDSC’s Communicable Disease Report [CDR],
7. Thacker SB, Stroup DF. Future directions for compre-
CDC’s Morbidity and Mortality Weekly Report hensive public health surveillance and health informa-
[MMWR], and selected surveillance data sets). This tion systems in the United States. Am J Epidemiol
network is the beginning of a larger international 1994;140:383-97.
network that will share data, exchange information, 8. Centers for Disease Control and Prevention. National
and improve public health. This larger network could Electronic Telecommunications System for Surveil-
link such systems as Salmnet, NETSS, and PHLIS lance—United States, MMWR 1990-1991;40:502-3.
to create a virtual on-line library of international 9. Bean NH, Martin SM, Bradford H. PHLIS: An elec-
surveillance data and information for public health. tronic system for reporting public health data from
remote sites. Am J Public Health 1992;82:1273-6.
With the diffusion of technology, internationally
networked electronic public health surveillance sys-
tems are gaining in importance. Their existence
clearly facilitates the rapid collection, analysis, and
dissemination of vital public health information and
promotes the establishment of effective interna-
tional public health policies.
T. Demetri Vacalis
Epidemiology Program Office
Centers for Disease Control and Prevention
Atlanta, Georgia, USA
Christopher L. R. Bartlett
PHLS Communicable Disease Surveillance Centre
London, England
Cheryl G. Shapiro
National Center for Infectious Diseases
Centers for Disease Control and Prevention
Atlanta, Georgia, USA

News and Notes
Communicable Diseases DxMONITOR:

Intelligence Compiling Veterinary
Communicable Diseases Intelligence (CDI) is a Diagnostic Laboratory Results
fortnightly publication of the Australian Department
The DxMONITOR is a collaborative effort be-
of Human Services and Health and the Communica-
tween the U.S. Department of Agriculture, Animal
ble Diseases Network of Australia and New Zealand.
and Plant Health Inspection Service’s Veterinary
The Network comprises representatives of the Aus-
Services (USDA:APHIS:VS), the American Associa-
tralian Department of Human Services and Health,
tion of Veterinary Laboratory Diagnosticians, and
the State and Territory health authorities, and other
organizations involved in communicable disease sur- the United States Animal Health Association. This
quarterly animal health report presents compiled
veillance and control from throughout the country. In
data from national animal disease control and eradi-
addition, there is a representative from New Zea-
cation programs (bovine and porcine brucellosis, bo-
land. It has fortnightly teleconferences and other
vine tuberculosis, porcine pseudorabies and equine
meetings to exchange information on emerging com-
infectious anemia); patterns of selected diseases
municable disease activity and to coordinate surveil-
lance and control activities. based on veterinary diagnostic laboratory data (bo-
vine leukosis; bovine bluetongue; bovine, ovine and
Each issue of CDI incorporates reports from Aus-
caprine paratuberculosis; equine arboviral encepha-
tralia’s national communicable diseases surveillance
litis; equine viral arteritis; porcine reproductive and
systems, including the National Notifiable Diseases
respiratory syndrome); data on selected etiologic
Surveillance System, the CDI Laboratory Reporting
agents associated with specific animal health events
Schemes, and the Australian Sentinel General Prac-
such as bovine abortion; global disease distribution
titioner Surveillance Network. Reports from the Na- (bovine spongiform encephalopathy); and notes from
tional Salmonella Surveillance Scheme, the veterinary diagnostic laboratories about unusual
Australian Gonococcal Surveillance Programme and
laboratory findings or new diagnostic procedures.
the National HIV, AIDS, and Tuberculosis Reporting
The DxMONITOR has contributed to a greater
Systems are also regularly included.
awareness of animal diseases in the United States.
CDI also publishes timely reports of communica-
Global trade agreements, the worldwide information
ble disease outbreaks and other articles dealing with
explosion, and increasing public concern over the
a wide range of subjects relevant to the surveillance
safety and quality of food have focused attention on
and control of communicable diseases in Australia. animal health. Compilation of veterinary diagnostic
Recently published items have reported, for exam- laboratory data is one component of the USDA:
ple, the first identification of endemically acquired
APHIS efforts to respond to these increased demands
hepatitis E in the Northern Territory of Australia, an
for animal health information through an integrated
outbreak of influenza in a nursing home, the
and coordinated monitoring and surveillance sys-
epidemiology of hepatitis A in South Australia, the
tem. Animal-health monitoring and disease surveil-
epidemiology of Barmah Forest virus disease in
lance concern not only animal health per se, but also
Western Australia, and the outbreak of respiratory interactions with the environment, animal welfare,
disease in humans and horses due to a previously
production practices and product wholesomeness
unrecognized paramyxovirus.
which impact animal health. The DxMONITOR is
mailed to all interested parties without charge and
CDI is available from is increasingly available through electronic dissemi-
The Editor nation channels. For more information or subscrip-
Communicable Diseases Intelligence tion, contact DxMonitor Animal Health Report, c/o
AIDS and Communicable Diseases Branch
Department of Human Services and Health Centers for Epidemiology and Animal Health,
GPO Box 9848 USDA:APHIS:VS, 555 S. Howes, Suite 200, Ft. Col-
Canberra ACT 2601 lins, CO 80521-2586; telephone 303-490-7800; e-mail
Australia DXMONITOR@aphis.ag.gov.
William Hueston
Helen Longbottom Centers for Epidemiology and Animal Health
Department of Human Services and Health Veterinary Services
Canberra ACT Australia APHIS, U.S. Department of Agriculture
Fort Collins, Colorado, USA

News and Notes
WHO Scientific Working Group and strategies to strengthen local, national, and
global surveillance. Participants included repre-
on Monitoring and Management sentatives from clinical medicine, public health, the
clinical laboratory, and the biomedical research
of Bacterial Resistance to arenas and from the pharmaceutical industry.
Antimicrobial Agents The Working Group formulated a series of recom-
mendations to address these issues at local, national,
Antibacterial resistance is a global clinical and and international levels. The recommendations
public health problem that has emerged with alarm- placed emphasis on enhanced surveillance of drug
ing rapidity in recent years and undoubtedly will resistance through usage of WHONET software, in-
increase in the near future. Resistant bacteria do not creased monitoring and improved usage of antimi-
respect national borders, and developments in the crobial drugs in human, veterinary, and animal
remote locations can have an impact throughout the husbandry settings, improved laboratory diagnostic
world. Resistance is a problem in the community as capacity, standardization and quality control of labo-
well as in health care settings, where transmission ratory methodology, professional and public educa-
of bacteria is greatly amplified, in both developed tion, development of new drugs and assessment of
and developing countries. Because multiple drug alternative therapeutic modalities, assessment of
resistance is a growing problem, physicians are now vaccine development and delivery priorities related
confronted with infections for which there is no effec- to antimicrobial resistance, better implementation
tive therapy. The morbidity, mortality, and financial of infection control measures, and evaluation of pre-
costs of such infections pose an increasing burden for vention strategies.
health care systems worldwide, but especially in
The Working Group plans to release its final re-
countries with limited resources.
port in the spring.
The Division of Communicable Diseases at the
World Health Organization, Geneva, Switzerland, Fred C. Tenover
recently convened a Scientific Working Group to National Center for Infectious Diseases
address the problem of drug-resistant bacterial in- Centers for Disease Control and Prevention
fections. From November 29 to December 2, 1994, Atlanta, Georgia, USA
participants from 23 countries reviewed and dis- James M. Hughes
cussed scientific data on the nature and costs of drug National Center for Infectious Diseases
resistance; recent national and global trends; ap- Centers for Disease Control and Prevention
proaches to limiting the emergence and spread of Atlanta, Georgia, USA
resistance in community and institutional settings;

Emerging Infectious Diseases

Uploaded by

Copyright:

Available Formats

Emerging Infectious Diseases

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Emerging Infectious Diseases

Uploaded by

Copyright:

Available Formats

Emerging

Vol. 1, No. 1, January–March 1995

Cat-Scratch Disease Russell Regnery and

Barmah Forest Virus Michael D. A. Lindsay

Shigella sonnei Infection J. A. Frost

Lyme Disease Richard C. Russell

Morbillivirus Pneumonia of Horses Keith Murray

Electronic Communication and T. Demetri Vacalis

Communicable Diseases Helen Longbottom

U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES

Emerging Infectious Diseases

Emerging Infectious Diseases

Electronic Access to Emerging Infectious Diseases

The early history of infectious diseases was characterized by sudden, unpredictable

“Nothing in the world of living things is permanently fixed.”

Vol. 1, No. 1 — January-March 1995 1 Emerging Infectious Diseases

Emerging Infectious Diseases 2 Vol. 1, No. 1 — January-March 1995

Getting Ahead of the Curve surveillance to be effective it must be specific. Con-

Table 1. Major Etiologic Agents, Infectious Diseases Identified Since 1973*

Vol. 1, No. 1 — January-March 1995 3 Emerging Infectious Diseases

Emerging Infectious Diseases 4 Vol. 1, No. 1 — January-March 1995

Vol. 1, No. 1 — January-March 1995 5 Emerging Infectious Diseases

Emerging Infectious Diseases 6 Vol. 1, No. 1 — January-March 1995

Factors in the Emergence of

“Emerging” infectious diseases can be defined as infections that have newly

Vol. 1, No. 1 — January-March 1995 7 Emerging Infectious Diseases

Emerging Infectious Diseases 8 Vol. 1, No. 1 — January-March 1995

Vol. 1, No. 1 — January-March 1995 9 Emerging Infectious Diseases

Table 2. Factors in infectious disease emergence*

Emerging Infectious Diseases 10 Vol. 1, No. 1 — January-March 1995

production system traditionally practiced in certain Changes in Human Demographics and

Vol. 1, No. 1 — January-March 1995 11 Emerging Infectious Diseases

International Travel and Commerce Technology and Industry

Emerging Infectious Diseases 12 Vol. 1, No. 1 — January-March 1995

Vol. 1, No. 1 — January-March 1995 13 Emerging Infectious Diseases

Emerging Infectious Diseases 14 Vol. 1, No. 1 — January-March 1995

Vol. 1, No. 1 — January-March 1995 15 Emerging Infectious Diseases

Unraveling Mysteries Associated with

Vol. 1, No. 1 — January-March 1995 16 Emerging Infectious Diseases

Enter Afipia felis Ultimately, the relationships between possible

Emerging Infectious Diseases 17 Vol. 1, No. 1 — January-March 1995

Vol. 1, No. 1 — January-March 1995 18 Emerging Infectious Diseases

Emerging Infectious Diseases 19 Vol. 1, No. 1 — January-March 1995

Vol. 1, No. 1 — January-March 1995 20 Emerging Infectious Diseases

Emerging Infectious Diseases 21 Vol. 1, No. 1 — January-March 1995

Emergence of Barmah Forest Virus

Vol. 1, No. 1 — January-March 1995 22 Emerging Infectious Diseases

Emerging Infectious Diseases 23 Vol. 1, No. 1 — January-March 1995

was isolated from Ae. camptorhyn-

Vol. 1, No. 1 — January-March 1995 24 Emerging Infectious Diseases

Emerging Infectious Diseases 25 Vol. 1, No. 1 — January-March 1995

An Outbreak of Shigella sonnei Infection

Vol. 1, No. 1 — January-March 1995 26 Emerging Infectious Diseases

Table 1. Shigella sonnei in England and Wales—Laboratory reports to CDSC

Week 1–20 Weeks 21–24 Weeks 1–24

Total Total Adults Women Total

Emerging Infectious Diseases 27 Vol. 1, No. 1 — January-March 1995

Table 2. Sh. sonnei case control study—Single variable matched analysis

Vol. 1, No. 1 — January-March 1995 28 Emerging Infectious Diseases